National Neonatal Audit Programme (NNAP) 2018 Annual Report on 2017 Data

National Neonatal RCPCH Audits Audit Programme (NNAP)

Annual report 2018 annual report on 2017 data

RCPCH Royal College of Paediatrics and Child Health

National Neonatal Audit Programme (NNAP) 2018 annual report on 2017 data

The National Neonatal Audit Programme is commissioned by the Healthcare Quality Improvement Partnership (HQIP) as part of the National Clinical Audit and Patient Outcomes Programme (NCAPOP).

HQIP is led by a consortium of the Academy of Medical Royal Colleges, the Royal College of Nursing and National Voices. Its aim is to promote quality improvement in patient outcomes, and in particular, to increase the impact that clinical audit, outcome review programmes and registries have on healthcare quality in England and Wales.

HQIP holds the contract to commission, manage and develop the National Clinical Audit and Patient Outcomes Programme (NCAPOP), comprising around 40 projects covering care provided to people with a wide range of medical, surgical and mental health conditions.

The programme is funded by NHS England, the Welsh Government and, with some individual projects, other devolved administrations and crown dependencies. www.hqip.org.uk/national-programmes

Published by RCPCH September 2018. The Royal College of Paediatrics and Child Health is a registered charity in England and Wales (1057744) and in Scotland (SCO38299)

National Neonatal Audit Programme 2018 report on 2017 data

Acknowledgements

The NNAP Project Board would like to thank all the doctors, nurses, administrators, data analysts and others who have given their time and effort to collect information for the audit and ensure its accuracy, and who have developed and carried out plans to improve the service they deliver. We would particularly like to thank the NNAP clinical leads in each unit, and the neonatal networks for their continued support.

We would also like to thank the people and organisations that work closely with the NNAP but are not represented on the Project Board or Methodology and Dataset Group, including the National Maternity and Perinatal Audit (NMPA), the Independent Advisory Group of the Healthcare Quality Improvement Partnership (HQIP), and the Neonatal Critical Care Clinical Reference Group at NHS England.

NNAP Project Board members

Professor Anne Greenough, Professor of Neonatology and Clinical Respiratory Physiology, Chair of the NNAP Project Board Dr Sam Oddie, Consultant Neonatologist, NNAP Clinical Lead Vanessa Attrell, Network Manager, South East Coast Neonatal Network Dr Lisa Barker, Consultant Neonatologist Zoe Chivers, Head of Services, Bliss (to July 2018) Ellen Hallsworth, Parent Representative (to June 2018) Wendy Hodgson, Neonatal Nurses Association Representative Dr Chris Kissack, Consultant Neonatologist, Scottish Representative Dr Helen Mactier, Consultant Neonatologist, British Association of Perinatal Medicine Representative Professor Neena Modi, Professor of Neonatal Medicine, Neonatal Data Analysis Unit, Imperial College London Gina Outram, Neonatal Nurses Association Representative (to April 2018) Dr Colin Peters, Consultant Neonatologist, Scottish Representative Dr Siddhartha Sen, Consultant Neonatologist, Welsh Representative Mirek Skrypak, Associate Director for Quality and Development, Healthcare Quality Improvement Partnership Patrick Tully, NNAP Parent Representative Sarah Walker, Project Manager, Healthcare Quality Improvement Partnership Professor Andrew Wilkinson, Emeritus Professor of Paediatrics, University of Oxford Dr Amy Young, Trainee in Neonatal Medicine

National Neonatal Audit Programme 2018 report on 2017 data

NNAP Methodology and Dataset Group members

Dr Sam Oddie, Consultant Neonatologist, Chair of the NNAP Methodology and Dataset Group Dr Julie-Claire Becher, Consultant Neonatologist Dr Kate Blake, Consultant Neonatologist Charlotte Bradford, Information Manager, Yorkshire and Humber Neonatal Network Dr Sanjeev Deshpande, Consultant Neonatologist Jacki Dopran, Senior Nurse Elizabeth Gallagher, Network Manager, Wales Neonatal Network Rebecca Lemin, Network Manager, South West Neonatal Network Dr Yinru Lim, Trainee in Neonatal Medicine Dr Kate Palmer, Consultant Neonatologist Mehali Patel, Senior Research Officer, Bliss Dr Oliver Rackham, Consultant Neonatologist, RCPCH Clinical Lead for Audit

NNAP Project Team*

Dr Sam Oddie, Consultant Neonatologist, NNAP Clinical Lead Rachel Winch, NNAP Audit Manager, RCPCH Mark Hannigan, Clinical Standards and Quality Improvement Manager, RCPCH Dr Marcia Philbin, Assistant Director of Research and Policy, RCPCH Melanie David-Feveck, Project Administrator, RCPCH (to March 2018) Karina Green, Project Administrator, RCPCH (from March 2018) Sandeepa Arora, NNAP Data Analyst, NDAU (from March 2018) Kayleigh Ougham, NNAP Data Analyst, NDAU Dr Nicholas Longford, NNAP Statistician, NDAU

*Note that members of the NNAP Project Team also sit on the Project Board and Methodology and Dataset Group.

National Neonatal Audit Programme 2018 report on 2017 data

Table of contents

2018 annual report on 2017 data ...... 2

Acknowledgements ...... 3

Table of contents ...... 5

Forewords ...... 6

Executive summary ...... 8

1. Introduction ...... 16

2. Key findings and recommendations ...... 23

3. Methods ...... 54

4. Case studies: how NNAP supports local quality improvement ...... 62

5. Full NNAP results ...... 70

Appendix A: Data completeness and unit level of participating units ...... 143

Appendix B: Recommendations by audience ...... 149

Appendix C: Glossary and abbreviations ...... 158

Appendix D: Useful resources ...... 160

Appendix E: Matching method of comparing outcomes for BPD ...... 163

Appendix F: “Pathogens” in the NNAP ...... 165

References ...... 167

National Neonatal Audit Programme 2018 report on 2017 data

Forewords

I am pleased to introduce the 11th annual report of the National Neonatal Audit Programme, which has been run by the Royal College of Paediatrics and Child Health since its inception in 2006.

The audit celebrates some key achievements in neonatal care this year; more very preterm babies are being admitted to neonatal units with a normal temperature and rates of magnesium sulphate administration to mothers at risk of very preterm birth have increased notably (from 53% with 17% missing data, to 64% with 8% missing data).

Variation, however, continues to exist between neonatal units and neonatal networks. There are clear opportunities for units and networks to use their NNAP data as a driver for quality improvement activities.

The audit achieves excellent engagement from the neonatal community and the high levels of data completeness achieved in most audit measures mean that the audit continues to be a robust source of information, enabling the neonatal community to make best use of their results to drive change.

The NNAP reports for the first time this year on new measures of parental partnership in neonatal care. The development of these new measures is a credit to the NNAP parent representatives Ellen Hallsworth and Patrick Tully and Bliss representative Zoe Chivers. Ellen and Zoe step down this year after several years of providing highly valuable insight to the NNAP. I thank them for their contribution to the audit.

Thank you also to those involved in writing this report and developing its recommendations, including the NNAP Project Board, Methodology and Dataset Group, the Project Team and Clinical Lead Professor Sam Oddie. Finally, I would like to thank the neonatal and wider perinatal teams for providing their essential input into the audit.

Professor Anne Greenough, Vice President Science and Research Chair of the NNAP Project Board Royal College of Paediatrics and Child Health

Follow @RCPCHtweets | www.rcpch.ac.uk

National Neonatal Audit Programme 2018 report on 2017 data

The NNAP expects, this year, to achieve full coverage of the 182 neonatal units in England, Wales and Scotland. Engagement in this national audit is accepted by many national bodies to be a key indicator of neonatal service quality. The measures used include processes (clinical and organisational) and outcomes and continue to address many different dimensions of healthcare quality. Refreshingly, several new ideas have been introduced in this report including measures relating to parental partnership in care, and place of birth of babies born at less than 27 weeks gestational age, which are known to influence important clinical outcomes. There is acknowledgement of the importance of linking maternity and neonatal data in collaboration with the National Maternity and Perinatal Audit (NMPA). A start is made in systematic analysis of rates of change in measures with time and their variation between units and networks. The publication of comparative data is not sufficient on its own to improve care and reduce variation in outcomes. This might partly explain the fact that, despite ongoing improvement in many longstanding measures, the pace of change has reduced for many with the persistence of marked regional variation. Approaches by the NNAP to improve access to comparative data through NNAP Online and encourage local quality improvement are to be welcomed. National initiatives in England and Scotland to train professionals in quality improvement methodology and to collaborate for improvement, if sustained, should add momentum. The NNAP has become a very important part of the landscape of UK neonatal care. In the context of current work to transform neonatal services in England and Scotland it is important that priority is given to quality improvement informed by national audit and benchmarking. Closer coordination of the work of the many agencies interested in the quality of neonatal care would help accelerate improvement.

Dr Gopi Menon, President British Association for Perinatal Medicine

Follow @BAPM_Official | www.bapm.org

National Neonatal Audit Programme 2018 report on 2017 data

Executive summary

Around 750,000 babies are born each year in England, Scotland and Wales, and of these nearly 105,000 or around 1 in 7, will require specialist neonatal care. The National Neonatal Audit Programme (NNAP) uses routinely collected data to support quality improvement in neonatal units of all types.

Established in 2006, the NNAP is commissioned by the Healthcare Quality Improvement Partnership (HQIP), funded by NHS England, the Scottish Government and the Welsh Government, and is delivered by the RCPCH. It forms part of the HQIP National Clinical Audit and Patient Outcomes Programme (NCAPOP). The RCPCH is currently contracted to deliver the NNAP from April 2017 to March 2021. For most audit measures, this report looks at care provided to babies with a final discharge from neonatal care between 1 January and 31 December 2017.

In addition to our existing audit measures, in 2017 the NNAP reported on new measures focussed on parental partnership in care; looking at minimising separation of mother and baby, and the presence of parents on consultant ward rounds. We hope that these measures will support neonatal units to achieve a partnership with parents in providing care. This year we also describe how many of the least mature babies are delivered in units best suited to care for them. Our final new measure describes, for the first time, how many babies develop necrotising enterocolitis.

National Neonatal Audit Programme 2018 report on 2017 data

Selected key findings and recommendations

These key findings were selected by consensus at the NNAP key findings workshop by a multidisciplinary and multiagency group of NNAP stakeholder representatives. For a full list of the key findings and recommendations for these, and other measures, see the key findings and recommendations section of the full report.

Antenatal magnesium sulphate

Giving magnesium sulphate to women who are at risk of delivering a preterm baby reduces the chance that their baby will develop cerebral palsy. The NNAP looks at whether mothers who delivered their baby at less than 30 weeks were given antenatal magnesium sulphate. Magnesium sulphate administration was much higher in 2017 than in 2016 (2017 – 64.1% of eligible mothers; 2016 – 53.3% of eligible mothers), reflecting rapid assimilation into practice of this aspect of NICE guidance, which is aimed at reducing cerebral palsy.

Selected recommendation:

To seek missed opportunities, and themes as to why magnesium was not given in line with NICE guidance, neonatal and maternity care staff in units with below average rates of administration should formally review records of babies born at less than 30 weeks where magnesium sulphate was not given to the mother.

National Neonatal Audit Programme 2018 report on 2017 data

Birth in a centre with a neonatal intensive care unit (NICU)

The NNAP looks at the proportion of babies born at less than 27 weeks gestational age who were born at a hospital with an on-site NICU. Babies who are born at less than 27 weeks gestational age are at high risk of death and serious illness. There is evidence that outcomes are improved if such immature babies are cared for in a NICU from birth. Three in four babies born less than 27 weeks gestational age were born at a hospital with an onsite NICU. Only two of 15 neonatal networks have more than 85% of these babies born within a hospital with an on-site NICU. Geographical size of network does not readily explain why more of some networks’ babies are delivered in centres with a NICU.

Selected recommendation:

Neonatal networks, maternity networks and local maternity systems in England, and their equivalent bodies in Wales and Scotland, which do not achieve delivery of 85% of babies less than 27 weeks in a hospital with an onsite NICU should review whether they have realistic plans to achieve improvements in this area, and develop plans if required.

National Neonatal Audit Programme 2018 report on 2017 data

Promoting normal temperature on admission for very preterm babies

More very preterm babies in England, Scotland and Wales are admitted with a normal temperature than has been recorded for other nations in the international literature.1,2,3 Sixty four percent of babies had a normal first temperature (36.5 to 37.5°C) measured within an hour of birth. This is an improvement in performance from recent years (2016 – 60.8%; 2015 – 58.1%) without an increase in hyperthermia – temperature above 37.5°C (2017 – 12.2%; 2016 – 12%). However there remains room for significant further improvement in the promotion of normothermia on admission to neonatal units for very preterm babies.

Selected recommendation:

Neonatal units should ensure that they have a care bundle in place, developed with multidisciplinary input, which mandates the use of evidence-based strategies to encourage admission normothermia of very preterm babies.

National Neonatal Audit Programme 2018 report on 2017 data

Necrotising enterocolitis

Necrotising enterocolitis (NEC) is a devastating illness which can follow preterm birth. One in twenty (5.6%; 428 of 8,228) babies born at less than 32 weeks gestational age developed necrotising enterocolitis (NEC). The NNAP uses a surveillance definition of NEC based on diagnosis at surgery, post-mortem or on the presence of clinical or radiographic signs.

Selected recommendation:

Neonatal units who validated their NEC data for 2017 should use NNAP Online to compare rates of NEC with other units, and use these comparisons to seek quality improvement opportunities.

National Neonatal Audit Programme 2018 report on 2017 data

Minimising separation of mothers and term and late preterm babies

The NNAP looks at the number of days that term and late preterm babies requiring low dependency care are separated from their mother. Variation exists in the average number of separation days between neonatal units and networks, for both term and late preterm babies. Findings for these two measures suggest that opportunities exist to reduce separation of mothers and term and late preterm babies by providing some neonatal care as transitional care.

Selected recommendation:

Neonatal units and trusts/health boards where transitional care cannot be delivered should work with their commissioners to develop the ability to deliver such care to minimise mother and baby separation, following the BAPM guidance A Framework for Neonatal Transitional Care.11

Full key findings by audit measure are available in chapter 2 of the main report.

National Neonatal Audit Programme 2018 report on 2017 data

Supporting quality improvement in neonatal care

The NNAP identifies areas for quality improvement in neonatal units in relation to the delivery and outcomes of care. The NNAP presents data to neonatal units and networks to facilitate quality improvement, alongside other initiatives in the following ways:

• NNAP Online is the audit’s interactive reporting tool. It is available at http://nnap.rcpch.ac.uk and can be used to compare performance at a unit, network and national level; supporting neonatal units and networks to share best practice and stimulate quality improvement activities. The NNAP also shares examples of good practice by showcasing case studies in the annual report, online and at our annual NNAP and Neonatal Data Analysis Unit (NDAU) Collaborator’s Meeting.

• NNAP unit results posters summarise a selection of the unit’s NNAP results which are most relevant to parents and carers. Neonatal units display the posters in a public area, and complete a second poster, which explains the actions they are taking in response to their audit results. Designed to be used alongside Your baby’s care (available at www.rcpch.ac.uk/nnap), our parents’ guide to the NNAP, the posters help to communicate the meaning and relevance of the audit results not only to parents, but to the wider team involved in caring for the baby and mother.

• NNAP quarterly reports support neonatal units and networks to monitor data quality and completeness and their ongoing performance throughout the data collection year. Quarterly reports enable units to review their provisional results at the end of the year before inclusion in the NNAP annual report.

The NNAP works closely with neonatal networks, adapting its measures and reporting to be responsive to the needs of the networks. The NNAP works closely with other national bodies and participates in several national initiatives, including the National Clinical Audit Benchmarking project (NCAB, a collaboration between HQIP and CQC), the Neonatal Peer Review Visit programme, NHS Choices and MyNHS Clinical Outcomes Publication and the Transparency and Open Data initiative.

National Neonatal Audit Programme 2018 report on 2017 data

Future developments in the NNAP

For the 2018 data year, we expect to achieve participation from all 15 neonatal units in Scotland, giving full participation across England, Wales and Scotland and would like to achieve UK wide participation in the future.

A new measure of neonatal nurse staffing levels will be reported for the 2018 data year, focussing on the proportion of shifts staffed according to relevant standards, and the number of additional shifts that would be required to meet those standards.

In 2017 and 2018 the NNAP has been collecting data on mortality. For a very few preterm babies (those who die before 44 weeks post menstrual age, in a non NNAP unit) this will require additional data entry, but for most cases, this reporting will be based on existing data flows.

National Neonatal Audit Programme 2018 report on 2017 data

1. Introduction

This is the 11th annual report of the National Neonatal Audit Programme (NNAP) delivered by the Royal College of Paediatrics and Child Health (RCPCH).

The NNAP supports professionals, families and commissioners to improve care provided by neonatal services who look after babies born too early, with a low birth weight or who have a medical condition requiring specialist treatment.

Since its conception as an England only audit, the NNAP has expanded to include Welsh units in 2012, and Scottish units in 2015.

The data presented in this report relate to the care provided to 104,183 babies discharged from neonatal care during the calendar year 1 January 2017 to 31 December 2017 in the 179 participating neonatal units (of a total of 182) in England, Wales and Scotland.

1.1. Aims

The aims of the audit are:

• To assess whether babies admitted to neonatal units in England, Scotland and Wales receive consistent high-quality care in relation to the NNAP audit measures that are aligned to a set of professionally agreed guidelines and standards.

• To identify areas for quality improvement in neonatal units in relation to the delivery and outcomes of care.

National Neonatal Audit Programme 2018 report on 2017 data

1.2. Scope

In 2017, the NNAP focussed on the following areas of neonatal care:

• Administering antenatal steroids • Administering antenatal magnesium sulphate • Birth in a centre with a neonatal intensive care unit (NICU) • Promoting normal temperature on admission for very preterm babies • Speaking with parents within 24 hours of admission • Involving parents in decision making through presence at consultant ward rounds • Screening on time for retinopathy of prematurity (ROP) • Measuring rates of infection • Measuring rates of bronchopulmonary dysplasia • Measuring rates of necrotising enterocolitis • Minimising inappropriate separation of mother and baby (term and late to moderate preterm) • Feeding breastmilk at discharge home • Carrying out follow-up assessment at two years of age • Measuring mortality rates

Full details of the 2017 audit measures are available in Chapter 5.

1.3. Future developments

The NNAP has responded to requests from stakeholders to include a measure of nurse staffing on neonatal units, and has introduced a three-part measure for 2018 data looking at; the proportion of nursing shifts numerically staffed according to guidelines and service specification, the proportion of shifts staffed according to guidelines and service specification for qualification in specialty, and the number of additional nursing shifts required to be worked to meet guidelines and service specification.

In 2017 and 2018 the NNAP has been collecting data on mortality in very preterm babies. For a very few preterm babies (those who die before 44 weeks post menstrual age, in a non NNAP unit) this will require additional data entry, but for most cases, this reporting will be based on existing data flows.

National Neonatal Audit Programme 2018 report on 2017 data

1.4. Quality improvement

The NNAP identifies areas for quality improvement in neonatal units in relation to the delivery and outcomes of care. The NNAP supports neonatal units and networks to achieve quality improvement in a number of ways, and collaborates with regional and national initiatives and groups.

NNAP quarterly reporting

Through the data year, the NNAP provide neonatal units and networks with summary reports of their cumulative results for each of the NNAP audit measures. These reports give units and networks the opportunity to identify areas for improvement in both data quality and performance as the audit year progresses and take early action if required.

Local action plans and case studies

Neonatal units and networks use their NNAP audit results to develop local action plans, focussing on one or several audit measures where they have identified opportunities for improvement. The NNAP shares case studies of quality improvement projects so that learning is shared across all units and networks.

See Chapter 4 for a selection of case studies from NNAP participant units. More case studies can be found on our web pages at: www.rcpch.ac.uk/nnap.

Collaborators meeting

The latest NNAP and NDAU Collaborators Meeting was held on 27 April 2018, and featured talks and presentations relating to the future direction of the NNAP, NICE guidance on developmental follow up of children and young people born preterm, and using audit data for quality improvement and local improvement case studies relating to NNAP measures.

Videos of the presentations can be accessed at: www.rcpch.ac.uk/resources/2018- nnapndau-collaborators-meeting-presentations

National Neonatal Audit Programme 2018 report on 2017 data

Working with neonatal networks, and relevant national bodies

Operational delivery networks (ODNs) are charged with supporting English health trusts to support delivery of care according to the service specification, to respond to relevant national priorities, and to work with commissioners to plan services that are responsive to local needs. Equivalent structures exist in Scotland and Wales. The NNAP considers and adapts its measures and reporting to be responsive to the needs of ODNs. This is achieved by presentation of findings to relevant meetings, and by representation of networks on NNAP committees. Other national bodies such as the NHS England clinical reference group, and the NHS Improvement ATAIN project are key partners to NNAP, and close contact is maintained with them, to ensure that measures and reporting are aligned and fit for purpose.

The National Clinical Audit Benchmarking (NCAB) project

The HQIP/CQC led NCAB project provides a visual snapshot of individual trust audit data set against individual national benchmarks. NCAB is a collaboration between HQIP and CQC, which aims to enhance the way inspectors, medical directors, local clinical audit teams and others engage, interact with and share clinical audit data.

NNAP participated in this project for the first time with 2016 data from units in England and Wales for a selected number of audit measures, and is participating again this year.

For more information about this project, please see: https://www.hqip.org.uk/national- programmes/clinical-audit-benchmarking/

Neonatal Peer Review Visit Programme in England

Through late 2017 and early 2018, the Quality Surveillance Team at NHS England have been running the Neonatal Peer Review Visit programme. The programme has used NNAP data to inform visits.

National Neonatal Audit Programme 2018 report on 2017 data

NHS Choices and MyNHS: Clinical Outcomes Publication

The NNAP participated in this initiative for the first time in 2016 by submitting data for six of the NNAP audit measures. The NNAP is participating again this year, with unit level data expected to be published on NHS Choices and MyNHS in December 2018.

For more information about this initiative, please see: https://www.hqip.org.uk/national-programmes/clinical-outcomes-publication/

Open data

The NNAP publishes data annually on data.gov.uk following publication of the annual report. Data are published at neonatal unit level for all hospitals participating in the audit in England, Scotland and Wales. These are the same data available on NNAP Online.

Data accessed via data.gov.uk should be interpreted alongside this annual report and NNAP Online.

1.5. Impact and engagement

Information for parents and families

Your baby’s care is a parent and carer’s guide to the NNAP and the audit results. Available in English and Welsh, it tells families what the audit is, what it aims to achieve, explains the results for key audit measures, and what families can do in response to the results. We ask units to make the booklet available to parents and carers in their unit.

Your baby’s care is available here: www.rcpch.ac.uk/nnap

The NNAP fair processing and parent information leaflet Your baby’s information, is available here: www.nnap.ac.uk/nnap.

National Neonatal Audit Programme 2018 report on 2017 data

NNAP unit results posters

Following a successful pilot scheme for 2015 data, and a UK-wide roll out for 2016 data, the NNAP continues to produce its NNAP unit results posters. The results poster, designed in collaboration with our parent, nurse and Bliss representatives, summarises a selection of the unit’s NNAP results which are most relevant to parents and cares. Neonatal units display the posters in a public area, and complete a second poster, which explains the actions they are taking in response to their audit results. Designed to be used alongside Your baby’s care, the posters help to communicate the meaning and relevance of the audit results not only to parents, but to the wider team involved in caring for the baby and mother.

NNAP unit results posters can be downloaded from NNAP Online http://nnap.rcpch.ac.uk.

User feedback and improving the audit

The NNAP works closely with participating units and networks to make sure the audit is fit for purpose and continues to be as useful as possible to support improvements in care. We gather ad hoc feedback about audit measures and outputs directly from unit and network staff, through representatives at our Project Board and Methodology and Dataset Group, and at our Collaborators Meeting. Annually we run a survey seeking feedback from audit stakeholders. We ask for feedback on how units are using the audit data, quarterly reporting, NNAP Online, the national annual report, the unit posters and Your baby’s care. We use the feedback to make improvements to the audit and the outputs we produce.

Collaboration with the National Maternity and Perinatal Audit

The National Maternity and Perinatal Audit (NMPA), commissioned by the Healthcare Quality Improvement Partnership in July 2016, is a national audit of NHS maternity services in England, Scotland and Wales. The NMPA is led by the Royal College of Obstetricians and Gynaecologists (RCOG), in partnership with the Royal College of Midwives (RCM), Royal College of Paediatrics and Child Health (RCPCH) and the London School of Hygiene and Tropical Medicine.

The RCPCH is represented on the NMPA Project Board by Professor Anne Greenough, and the NNAP is represented on the NMPA Clinical Reference Group by Dr Sam Oddie.

National Neonatal Audit Programme 2018 report on 2017 data

The first clinical report of the NMPA was published in November 2017. The findings and recommendations in the report are wide reaching and of relevance to the neonatal community. In a whole system approach, measures included in the NMPA should be considered alongside the NNAP measures of neonatal care, and perinatal mortality reporting through Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK (MBRRACE-UK).

The NNAP team has been supporting the NMPA in the development of their latest project, a feasibility study aiming to link maternity and neonatal care records. We look forward to ongoing collaboration with the NMPA team in years to come.

You can find out more about the NMPA here: http://www.maternityaudit.org.uk.

1.6. NNAP Online

NNAP Online is the audit’s interactive reporting tool. It is available at http://nnap.rcpch.ac.uk. NNAP Online provides unrestricted access to all NNAP results at an individual unit level and for each measure.

NNAP Online can be used to compare performance at a unit, network and national level. It enables units to compare themselves against other units of the same designation and enables units and networks to share best practice and stimulate quality improvement activities.

NNAP Online includes:

• Neonatal unit and neonatal network annual summary reports • Graphical outputs for units and networks • Interactive outlier caterpillar plots • Posters of unit-level results • Encephalopathy results (at trust or health board level)

National Neonatal Audit Programme 2018 report on 2017 data

2. Key findings and recommendations

The NNAP brings together a multidisciplinary group of representatives identify key findings and to translate the key findings and results of the audit into a set of recommendations that can be acted upon to improve neonatal care. The recommendations are made to support the existing goals and priorities of neonatal and perinatal services, and are targeted to the audience with the ability to action the recommendation.

Recommendations are designed to be specific to each audit measure, however there are a number of recommendations for neonatal units relating to quality improvement activities across all NNAP measures:

1. Neonatal units should display their NNAP results poster and the accompanying poster describing the ongoing relevant quality improvement activities that the unit is making, in public and professional facing areas of the neonatal unit.

2. Neonatal units should use NNAP Online to identify quality improvement opportunities relevant to them, and to identify partner units with results they wish to emulate.

3. Neonatal units should ensure they have adequate processes for the timely capture of information for quality improvement, and build in regular review processes to measure their improvement progress.

Full recommendations can be found by audit measure in this chapter, and by audience in Appendix B: Recommendations by audience in the full report.

National Neonatal Audit Programme 2018 report on 2017 data

2.1. Antenatal steroids

Babies born at less than 35 weeks gestational age sometimes have breathing difficulties in the first few days after they are born. Antenatal steroids are a powerful health intervention, given to mothers by obstetricians and midwives before delivery of a preterm baby to help reduce breathing difficulties (respiratory distress syndrome) and make other serious complications such as bleeding into the brain less likely.

Key findings

• All neonatal networks and 84.9% (152 of 179) of units are meeting the NNAP standard of 85% of eligible mothers receiving at least one dose of antenatal steroids (Table 5.1.2., page 74 and NNAP Online).

• 88.6% of women who delivered a baby between 24 and 34 weeks’ gestational age received at least one dose of antenatal steroids, 1.5% more than in 2016. Units vary in their recorded use of antenatal steroids, with rates from 64.3% to 100%. One unit (Queens Hospital, Romford) had significantly lower use of antenatal steroids than other units in 2017; this unit was also a low outlier in 2016 data (NNAP Online).

• Two units (St Mary’s Hospital, Manchester and Birmingham Heartlands Hospital) improved their antenatal steroids coverage exceptionally over the period 2015 to 2017, from 73.4% to 90.3% and from 79.6% to 94.0% respectively (NNAP Online).

• Two units (Gloucestershire Royal Hospital and John Radcliffe Hospital, Oxford) demonstrated an exceptional decline in use of antenatal steroids from 2015 to 2017 (95.8% to 86.5% and 91.9% to 85.3% respectively) whilst remaining above the NNAP standard of 85%. This suggests that high performing units may need to retain focus on timely use of antenatal steroids in very preterm babies (NNAP Online).

National Neonatal Audit Programme 2018 report on 2017 data

Recommendations

4. Perinatal services (maternity and neonatal staff) should regard their rates of antenatal steroid administration as a key measure of the achievements of their clinical care. To identify quality improvement opportunities, neonatal and maternity care staff should formally review records of babies born at less than 35 weeks admitted for neonatal care where antenatal steroids were not given to the mother as part of their assurance with respect to NICE guidance.18

5. Neonatal networks should review administration rates of antenatal steroids in their units on a quarterly basis, identify any quality improvement opportunities and support units to achieve the best possible neonatal outcomes.

6. The NNAP and the National Maternity and Perinatal Audit should consider whether antenatal steroid administration could be more appropriately audited as part of the National Maternity and Perinatal audit from 2019 onwards.

7. Those responsible for defining national maternity datasets (NHS Digital in England) should ensure that antenatal steroid administration is captured as part of routine maternity data.

Full 2017 results for Antenatal steroids and a description of the measure are found on page 71.

National Neonatal Audit Programme 2018 report on 2017 data

2.2. Antenatal magnesium sulphate

Giving magnesium sulphate to women who are at risk of delivering a preterm baby reduces by 32% the chance that their baby will develop cerebral palsy.4 The NICE quality standard Preterm Labour and Birth recommends that all women that may deliver their baby at less than 30 weeks gestational age are offered magnesium sulphate where possible.18

Key findings

• Magnesium sulphate administration was much higher in 2017 than in 2016 (2017 – 64.1% of eligible mothers; 2016 – 53.3% of eligible mothers), reflecting rapid assimilation into practice of this aspect of NICE guidance, which is aimed at reducing cerebral palsy.18 Twenty-two units had administration rates of 80% or more of eligible women (Table 5.1.3, page 75 and NNAP Online).

• Marked variation is evident at network level, with the best performing network having a rate of administration more than one and a half times (78.8%) that of the lowest performing network (49.0%) (Table 5.2.2, page 79).

• The performance of three networks (Thames Valley and Wessex, South West, North West) is significantly better than average, and that of three networks (Staffordshire, Shropshire and Black Country, Midlands South West, Yorkshire and Humber) is significantly lower than average. Some networks (e.g. North West 2016 – 48%; 2017 – 70.3%) have made very rapid improvements to their performance (NNAP Online).

• John Radcliffe Hospital, Oxford reports exceptionally high rates of administration (91.2% of 102 women; national rate 64.1%), demonstrating that high rates of administration are achievable. Elsewhere, administration rates varied greatly

National Neonatal Audit Programme 2018 report on 2017 data

between units (from 0% to 100%) suggesting that uptake of this important treatment to prevent cerebral palsy is not yet optimal (NNAP Online).

Recommendations

8. Neonatal units with below average rates of magnesium sulphate administration should identify comparable units to their own, that have higher rates of antenatal magnesium sulphate administration using NNAP Online. Working collaboratively with maternity staff, they should use quality improvement methodology and programmes to improve rates of administration in their hospitals.

9. To seek missed opportunities, and themes as to why magnesium was not given in line with NICE guidance,18 neonatal and maternity care staff in units with below average rates of administration should formally review records of babies born at less than 30 weeks where magnesium sulphate was not given to the mother.

10. Neonatal units with significant levels of missing data should take steps to address this in collaboration with maternity care staff.

11. Neonatal networks, maternity networks and local maternity systems with below average rates of administration, or low rates of improvement review administration rates of magnesium sulphate in their units on a quarterly basis, identify any quality improvement opportunities and support units to achieve the best possible neonatal outcomes.

12. The NNAP and NMPA should explore the feasibility of reporting antenatal magnesium administration in NMPA.

Full 2017 results for Antenatal magnesium sulphate and a description of the measure are found on page 76.

Improving antenatal magnesium sulphate administration: a case study The neonatal unit at Watford General Hospital took a multidisciplinary approach to increase their rates of magnesium sulphate administration. Their top tips are:

• Identify an appropriate maternity-neonatal forum to share NNAP MgSO4 data • Engage stakeholders and frontline champions in maternity and neonates • Use live NNAP dashboard on BadgerNet to generate live run charts • Partner with parents for improvement

Find the full case study in chapter 4.1 and online at: www.rcpch.ac.uk/nnap

National Neonatal Audit Programme 2018 report on 2017 data

2.3. Birth in a centre with a neonatal intensive care unit (NICU)

Babies who are born at less than 27 weeks gestational age are at high risk of death and serious illness. National recommendations in England5 state that neonatal networks should aim to configure and deliver services to increase the proportion of babies at this gestational age that are delivered in a hospital with a neonatal intensive care unit (NICU) on site. This is because there is evidence that outcomes are improved if such immature babies are cared for in a NICU from birth.

Key findings

• Only two of 15 neonatal networks have more than 85% of babies born at less than 27 weeks gestational age within a hospital with an on-site NICU. Geographical size of network does not readily explain why more of some networks’ babies are delivered in centres with a NICU (Table 5.3.1, page 82).

• Networks varied by important margins in how they performed on this measure. One neonatal network, East of England ODN, has a low rate of delivery of babies born at less than 27 weeks gestational age in a hospital with an on-site NICU making it a low outlier at more than three standard deviations (SD) from the national rate (East of England ODN – 46.4%, national rate - 73.9%). Three other neonatal networks had rates that were more than two standard deviations above the national rate (NNAP Online).

National Neonatal Audit Programme 2018 report on 2017 data

Recommendations

13. Neonatal networks should facilitate local and network review of all cases where babies of less than 27 weeks gestational age deliver in a hospital without a NICU with the aim of identifying and sharing opportunities to increase the rate of delivery in hospitals with an onsite NICU.

14. Neonatal networks, maternity networks and local maternity systems in England, and their equivalent bodies in Wales and Scotland, which do not achieve delivery of 85% of babies less than 27 weeks in a hospital with an onsite NICU should review whether they have realistic plans to achieve improvements in this area, and develop plans if required.

Full 2017 results for Birth in a centre with a NICU and a description of the measure are found on page 81.

National Neonatal Audit Programme 2018 report on 2017 data

2.4. Promoting normal temperature on admission for very preterm babies

Low admission temperature has been associated with an increased risk of illness and death in preterm babies. Low temperature (or hypothermia) is a preventable condition in vulnerable newborn babies. Staff on the neonatal unit need to know if a baby is too cold so they can take appropriate action. This NNAP measure looks at how successful neonatal units are at achieving a normal first temperature (36.5°C to 37.5°C inclusive) within an hour of birth in very preterm babies (less than 32 weeks gestational age).

Key findings

• More very preterm babies in England, Scotland and Wales are admitted with a normal temperature than has been recorded for other nations in the international literature.2, 3,6 Sixty-four percent of babies had a normal first temperature (36.5 to 37.5°C) measured within an hour of birth. This is an improvement in performance from recent years (2016 – 61%; 2015 – 58%) without an increase in hyperthermia – temperature above 37.5°C (2017 – 12.2%; 2016 – 12%). However there remains room for significant further improvement in the promotion of normothermia on admission to neonatal units for very preterm babies (Table 5.4.1, page 85, Table 5.4.3, page 90).

• Temperature was measured and recorded in 99.7% (7,997 of 8,019) of babies (Table 5.4.1, page 85).

• Seven units (Bradford Royal Infirmary, Leeds Neonatal Service, John Radcliffe Hospital, Oxford, Princess Anne Hospital, Southampton, Rosie Maternity Hospital, Cambridge, The Royal London Hospital, and Queen Alexandra Hospital, Portsmouth) were positive outliers and achieved rates of documented within range temperatures of 71.1% to 87.5% (NNAP Online).

National Neonatal Audit Programme 2018 report on 2017 data

• Some units (Royal Victoria Infirmary, Newcastle, The Royal London Hospital) have increased their performance by 30% or more in two years, against a national average improvement of 5.4% over the same time (NNAP Online).

• While all networks have improved their performance between 2015 and 2017, some had a lower rate of improvement, e.g. Midlands South West (2017 – 50.6%; 2016 – 48%; 2015 – 42%) (NNAP Online).

• Just over one in 20 (5.6%) babies born before 32 weeks gestational age was markedly hypothermic (temperature less than 36.0°C) on admission to the neonatal unit. In 2015, 1 in 12 babies (8.3%) were markedly hypothermic (Table 5.4.3, page 90).

Recommendations

15. Neonatal units should report all cases where the admission temperature of a very preterm baby is below 36.00C using local risk reporting mechanisms, and consider a policy of reporting all babies with admission temperature below 36.50C.

16. Neonatal units should ensure that they have a care bundle in place, developed with multidisciplinary input, which mandates the use of evidence-based strategies to encourage admission normothermia of very preterm babies.

Full 2017 results for Promoting normal temperature on admission for very preterm babies and a description of the measure are found on page 84.

Improving admission temperature: a case study Sheffield Teaching Hospitals NHS Foundation Trust put in place a care bundle for thermal care at delivery. The approach included:

- Education and awareness raising through induction and huddles

- Monthly admission temperature tracking

- Exception reporting of admission temperatures below 36.00C

Find the full case study online at: https://www.rcpch.ac.uk/resources/2018-nnapndau- collaborators-meeting-presentations

National Neonatal Audit Programme 2018 report on 2017 data

2.5. Parental consultation within 24 hours of admission

This measure of care looks at whether parents have been spoken to by a senior member of the neonatal team within the first 24 hours of their baby being admitted. It applies for all babies who require care on a neonatal unit. It is important that families understand and are involved in the care of their baby. This first consultation provides an opportunity for the senior staff member to meet the parents, listen to their concerns, explain how their baby is being cared for and respond to any questions.

Key findings

• Rates of parental consultation are stable, with 40% units (72 of 179) having performance of over 98%, and 29 units having a senior member of the neonatal team speaking to all parents within 24 hours of admission. The national rate was 94.6% (2016 - 94%; 2015 - 91.9%) (Table 5.5.3, page 94, NNAP Online).

• Nineteen units had performance that was low outlying. However low outlying units performed better than in 2016 – for example just three units (James Paget Hospital, Nevill Hall Hospital and St Michael’s Hospital) had consultation rates of below 80% (2016 - 26 units) (NNAP Online).

• Most (12/15) networks improved their performance from 2015 to 2017. The worst performing network in 2015 (Midlands South West) was the most improved network between 2015 and 2017 (2015 - 70%; 2017 – 91%) (NNAP Online).

National Neonatal Audit Programme 2018 report on 2017 data

Recommendations

17. Neonatal units should regularly review the reasons why timely parental consultations did not occur. They should look for themes among the reasons, provide regular feedback to neonatal staff, and put processes in place to strengthen their support of parental partnership in care.

18. Neonatal units should ensure that parents are aware of the standard, for example as part of a welcome pack or signage in the neonatal unit.

19. Neonatal units with poorer data completeness should review and improve their documentation process. For example, by use of a dedicated notes sheet or a document in electronic records to record parental consultations.

Full 2017 results for Parental consultation within 24 hours of admission and a description of the measure are found on page 91.

Improving consultation with parents: a case study The Royal Oldham Hospital sparked the competitive spirit of their neonatal consultants to improve their rates of consultation with parents within 24 hours of admission.

A weekly run chart presented at communications meetings has resulted in a change in practice and sustained improvement.

Find the full case study in chapter 4.2 and online at: www.rcpch.ac.uk/nnap

National Neonatal Audit Programme 2018 report on 2017 data

2.6. Parental presence at consultant ward rounds

Neonatal intensive care is very stressful for babies and parents. Professionals, parents’ advocates, and parents agree that parental partnership in care is supported by including parents in consultant ward rounds, which will occur regularly on neonatal units. For 2017 this measure seeks to identify the proportion of admissions where parents were present on a consultant ward round on at least one occasion during a baby’s stay.

Key findings

• For 74.3% of neonatal stays parents were documented as having attended a consultant ward round at least once. This figure was 87.5% for neonatal admissions longer than 28 days, but the percentage of missing data was considerably higher for shorter stays (16% for stays of more than 7 days, less than 1% for stays longer than 28 days) (Table 5.6.1, page 97).

• In 10.6% of admissions, the reason given for parents not attending a consultant ward round was that no consultant ward round occurred. This is a new data measure for 2017 and it may be that differing interpretations are being applied to the term “consultant ward round” (Table 5.6.1, page 97).

• Although the validity of the findings may be affected by data completeness, the results suggest variation between neonatal networks in how frequently parents attend ward rounds, and can be involved in planning care. Attendance at any consultant ward round for stays longer than 28 days ranged from 76-98%, indicating that the model of parental partnership in care may yet be differentially adopted in UK neonatal care (Table 5.6.2, page 98).

National Neonatal Audit Programme 2018 report on 2017 data

Recommendations

20. Neonatal units with poorer data completeness should review and improve their documentation process to ensure that all instances of parental presence on the ward round are recorded.

21. Neonatal units should work with local parent representatives to look at ways to improve the attendance of parents on the ward round and parental involvement in decision making. Neonatal units should refer to the BAPM Neonatal Service Quality Indicators7 and the Bliss Baby Charter8 for guidance.

Full 2017 results for Parental presence on consultant ward rounds and a description of the measure are found on page 95.

National Neonatal Audit Programme 2018 report on 2017 data

2.7. On-time screening for retinopathy of prematurity

Babies born very early or with a very low birth weight are at risk of retinopathy of prematurity (ROP). This condition affects the development of the blood vessels in the back of the eye. ROP can lead to loss of vision, but this is usually prevented by timely treatment. Therefore, screening babies for ROP at the right time is important to help babies have the best vision in the future. A national guideline indicates when screening should be done, and this measure reports on how successful neonatal services are in achieving ‘on time’ screening.29

Key findings

• 94.4% of babies had on-time ROP screening (2016 data 94.2%), which means that the pace of improvement in this area is slowing (Table 5.7.1, page 104).

• 98.1% of eligible babies had ROP screening reported at any time (2016 data 98.4%).

• Most babies who were not screened had birthweights or gestational age just below the relevant criteria for screening – for example mature growth restricted babies, or larger immature babies. However, 13 babies with gestational age less than 30 weeks appear not have been screened at any time.

• As in previous years, more than one in eight babies had their first ROP screen after discharge, reflecting increased early discharge home of preterm and growth restricted babies since the national guidelines were written.29

• Forty-seven units (26.3%) reported screening all their eligible babies on time, including 13 large (more than 50 eligible babies) units. Eighty-one units (45.3%) screened less than 95% of eligible babies on time.

National Neonatal Audit Programme 2018 report on 2017 data

• Nine units were identified as low outliers – of whom three were also identified as low outliers in the 2016 data (James Cook University Hospital, Middlesbrough, Newham General Hospital, Birmingham Women’s Hospital) (NNAP Online).

• Network level performance varies considerably, suggesting it may be possible for networks to improve their practice by comparing organisational and clinical arrangements to, and basing improvements on, networks with good performance. One network (North West) is identified for its excellent performance (97.4% on time screening of 1046 babies) (Table 5.7.2, page 105).

• Three networks - Wales, Scotland, Trent Perinatal and Central Newborn Neonatal ODN - showed the most improvement between 2015 and 2017. Only the Northern Neonatal network did not show improvement – in 2017 it screened less than 9 out of 10 of its babies on time (86% on time screening of 342 babies), and in 2017 is a low outlier (NNAP Online).

• At unit level, some units can be identified as having exceptional improvements in their performance between 2015-2017 (Royal Victoria Infirmary, Newcastle, Queens Hospital, Romford, Walsall Manor Hospital). Two other units have declining rates of ROP screening between 2015-2017 that make them outliers (Whipps Cross University Hospital, Rosie Maternity Hospital, Cambridge) (NNAP Online).

National Neonatal Audit Programme 2018 report on 2017 data

Recommendations

22. Neonatal units and ophthalmologists should target quality improvement in their organisational, administrative and clinical processes at those babies whose birthweights and gestations are just inside the criteria for screening, because these babies constitute the majority of those not screened.

23. Neonatal units with low outlier status, and especially those who have been recurrently identified as such, should urgently review their clinical, administrative and organisational arrangements, and keep them under detailed regular review to optimise retinopathy screening and treatment outcomes.

24. Neonatal units should, as part of a formal local risk incident investigation, formally review their clinical, organisational and administrative pathways in discussion with their ophthalmology colleagues when cases are screened late, or not at all.

25. Neonatal units should clearly describe to parents, prior to the opening of the screening window, but after the first week of life, the need for ROP screening using an individualised written resource which sets out for the parents the anticipated date of first screening for their baby. If their baby is due to be screened after being discharged from the unit, neonatal staff should ensure that parents are aware of the importance of attending the appointment.

26. Neonatal networks with higher rates of failure to screen on time (for example over 2.5%) should seek to understand the reasons for this failure and address this with any units concerned.

27. Guideline developers should take the successful deployment of on time post discharge screening into account when describing appropriate clinical practice for ROP screening.

Full 2017 results for On-time screening for ROP and a description of the measure are found on page 102.

National Neonatal Audit Programme 2018 report on 2017 data

2.8. Encephalopathy

Encephalopathy is a brain illness. Encephalopathic babies have impaired consciousness and often have seizures. Encephalopathy in newborn babies has a variety of causes. Encephalopathy most commonly occurs in babies who are born at or near term and who appear to have got into difficulty during labour or delivery. It is important that hospitals gain understanding of their rates of encephalopathy in newborn babies to identify opportunities to improve midwifery and obstetric practice.

Key findings

• NNAP present rates of encephalopathy by trust or health board for the years 2014- 2016 inclusive, using number of live births as a denominator. The rates of encephalopathy presented are not risk adjusted according to maternal or obstetric characteristics, and so cannot be seen as indicative of the quality of antenatal or intrapartum care (NNAP Online).

Recommendations

28. Neonatal units should ensure that all cases of encephalopathy identified by the criteria used by the NNAP have been reviewed by a suitable multidisciplinary group to look for modifiable factors, in accordance with the approach taken in the Royal College of Obstetricians and Gynaecologists’ “Each Baby Counts” programme.9

Full 2017 results for Encephalopathy and a description of the measure are found on page 107.

National Neonatal Audit Programme 2018 report on 2017 data

2.9. Measuring rates of infection

Sick and premature babies are prone to infection with germs including some that are normally harmless to healthy people. Infections can lengthen the stay in the neonatal unit and may worsen the long term developmental outlook for babies.10 Neonatal unit staff and parents can reduce the risk of infection by following good infection prevention and control practice.

The NNAP focusses on reporting measures of bloodstream infection. To look for infection in babies, neonatal staff usually take blood cultures to check whether bacteria are present in the blood. They may also take a sample of cerebrospinal fluid (CSF). For 2017 data NNAP reports rates of blood cultures positive for bacteria, fungi or yeasts, and two different measures of bloodstream infection that occurs on the same day as a central line is present.

Key findings

Bloodstream infection

• 74 (41% of 179) neonatal units have provided assurance that 100% of positive blood cultures reported in their units have been submitted to the audit. This compares favourably to 2016 when 25 units could provide this assurance. This means that some inter-unit comparisons of infections can be made, for measures that do not depend on definitions of infection that depend on so called “symptoms and signs”. For units that have not offered this assurance the unknown number of unreported positive cultures renders inter-unit comparisons unreliable.

• For babies born at less than 28 weeks gestational age, confirmed infection rates appear high. 2563 babies had 521 growths of a pathogen, confirming the clinical importance of infection in this patient group (Table 5.9.2, page 111).

Full 2017 results for Bloodstream infection and a description of the measure are found on page 109.

National Neonatal Audit Programme 2018 report on 2017 data

Quality improvement surveillance definition (QISD): Central line associated bloodstream infection

• 74 (41% of 179) NNAP units have reported complete data entry of all positive blood cultures, making their own infection rates suitable for comparison to other neonatal units with complete data entry (see NNAP Online). In 2016 just 25 units reported complete data entry.

• In the 74 units with complete data entry, central line associated bloodstream infection occurred in 8.17 babies of less than 32 weeks gestational age per 1000 line days and in 2.84 babies of greater than or equal to 32 weeks gestational age per 1000 line days. This is more than that reported in units without complete data entry (5.85 per 1000 line days for babies less than 32 weeks gestational age, 2.15 per 1000 line days for babies greater than or equal to 32 weeks gestational age). The likely explanation is incomplete data entry in the units without known complete data entry (Table 5.10.1, page 114).

Full 2017 results for QISD: Central line associated bloodstream infection and a description of the measure are found on page 113.

National Neonatal Audit Programme 2018 report on 2017 data

Recommendations

29. Neonatal units should enter every blood culture result that is positive for any bacterial or fungal growth (including potential contaminants) and use a regular communication channel with their laboratory services to assure themselves and other NNAP audit users that their data entry is complete.

30. Neonatal units with complete entry of positive blood cultures and above average rates of bloodstream infection with a known pathogen in babies of less than 32 weeks gestational age should consider identifying suitable partner units from NNAP Online with lower rates of infection and comparing their infection reduction strategies to seek quality improvement opportunities.

31. Neonatal units with complete entry of positive blood cultures and above average rates of bloodstream infection with central line use as measured by the CRG / NNAP quality improvement surveillance definition of central line associated bloodstream infection (QISD CLABSI) should consider identifying suitable partner units from NNAP Online, and comparing their infection reduction strategies to seek quality improvement opportunities.

32. The NNAP should continue to seek to achieve linkage between other infection surveillance systems and the National Neonatal Research Database (NNRD) to report meaningful data about bloodstream infection.

Measuring infection: a case study The Jessop Wing NICU, Sheffield Teaching Hospitals NHS Foundation Trust, implemented a three-step approach to recording their central line associated bloodstream infection rates:

1. Monthly e-mail from microbiology to named neonatal consultant with all positive blood culture results (between 2 and 8 per month).

2. Named consultant ensures results are recorded within the BadgerNet system.

3. Data team use BadgerNet tools to check any drug/line discrepancies to minimise line day errors.

Find the full case study online at: https://www.rcpch.ac.uk/resources/2018-nnapndau- collaborators-meeting-presentations

National Neonatal Audit Programme 2018 report on 2017 data

2.10. Bronchopulmonary dysplasia (BPD)

Babies born preterm often don’t have fully developed lungs and may require support with their breathing from a ventilator or other device. Simply being born early can cause some ongoing breathing difficulty. Being on a ventilator can cause damage to the lungs, exacerbate breathing problems later in life and put babies at risk of chest infections. This condition is known as bronchopulmonary dysplasia (BPD), also called chronic lung disease. NNAP reports on the proportion of babies born very early who are still receiving help with their breathing or extra oxygen four weeks before their due date.

Variations in rates of BPD might reflect different management or could reflect the way that neonatal units use oxygen in most mature babies.

Key findings

• Thirty one percent of admitted babies born at less than 32 weeks met the surveillance definition for bronchopulmonary dysplasia (receiving respiratory support and/or supplemental oxygen at 36 weeks’ postmenstrual age) in the period 2015-2017 (Table 5.12.1, page 119).

• Marked variation exists at unit level in reported rates of both BPD alone and the composite measure of death or BPD. Among NICUs reported rates of BPD alone vary from below 25% to above 50% (mean 37.1%). Case mix characteristics explain much of, but not all, the variation in rates of BPD or death (NNAP Online).

• Important variation also exists between neonatal networks in their rates of both BPD and the combined outcome of death or BPD. These differences in rate are not wholly explained by differences in baseline characteristics of the cases. Rates of BPD or death are lower in three networks (East of England, South East Coast, Thames Valley and Wessex), and higher in two networks (Scotland and North West) than would be

National Neonatal Audit Programme 2018 report on 2017 data

expected if the babies cared for in these networks in 2015-2017 had been cared for in other units. A description of the novel matching approach used to compare networks and units is explained on page 58 and in Appendix E: Matching method of comparing outcomes for BPD.

Recommendations

33. Neonatal units with a positive treatment effect should consider examining the practice of neonatal units with a negative treatment effect to identify potential modifiable factors in their neonatal care which might influence rates of BPD.

34. Neonatal networks with a positive treatment effect should consider examining the practice of networks with a negative treatment effect to identify potential modifiable factors in their neonatal care which might influence rates of BPD.

35. When the NICE guidance on specialist neonatal respiratory care for babies born preterm is published, neonatal networks and neonatal units should review their policies to ensure that saturation targets are in line with best practice recommendations.

Full 2017 results for Bronchopulmonary dysplasia and a description of the measure are found on page 117.

National Neonatal Audit Programme 2018 report on 2017 data

2.11. Necrotising enterocolitis

Necrotising enterocolitis (NEC) is a devastating illness which can follow preterm birth. Bowel inflammation prevents milk feeding, surgery may be needed and babies who develop NEC typically stay in hospital for a long time. Rates of mortality in babies with NEC are high, at over 20%. Babies who survive NEC can have developmental problems when they are older.

Key findings

• One in twenty (5.6%; 428 of 8,228) babies born at less than 32 weeks gestational age developed necrotising enterocolitis (NEC). The NNAP uses a surveillance definition of NEC based on diagnosis at surgery, post-mortem or on the presence of clinical or radiographic signs (Table 5.13.1, page 125).

• 84 units (of which 18 cared for 30 or more babies, accounting for 1438 of 8228 eligible babies) reported no cases of NEC at all, which may be an unexpected finding (NNAP Online).

• Rates of NEC in the 78 units who had validated their NEC data for 2017 were 0.5% lower than rates in units who had not formally confirmed the local validation of their data.

• Most (5469 of 8228) of the very preterm (less than 32 weeks gestational age) babies eligible for this measure were in NICUs by 48 hours of age, but 2542 babies were nursed in Local Neonatal Units. Of the eligible babies in Local Neonatal Units, 3% (68) developed NEC which represents 1 in 6 of all babies who developed NEC. This emphasises the importance of delivering quality improvement opportunities at units of all levels caring for babies who are at risk of developing NEC (Table 5.13.1, page 125).

National Neonatal Audit Programme 2018 report on 2017 data

Reported levels of NEC by neonatal network appeared to vary threefold (South West 3.2% – South London 10.5%) (Table 5.13.2, page 126).

• Data was known to be missing for 7.3% (600 of 8228) of eligible babies. Of these 110 of a total of 393 deaths had no data on the occurrence of NEC (Table 5.13.1, page 125).

Recommendations

36. Neonatal units who validated their NEC data for 2017 should use NNAP Online to compare rates of NEC with other units, and use these comparisons to seek quality improvement opportunities.

37. Neonatal units should ensure that they will be able to validate their NEC data entry for the 2018 data year.

38. Neonatal networks should support neonatal units providing all levels of care to undertake quality improvement activities relating to NEC.

39. The NNAP should consider increasing the time period for reporting NEC, to a rolling period of three years to maximise the discriminatory power of this measure.

40. The NNAP should consider reporting a combined outcome of NEC or death from the 2019 data year, and should consider applying a matching approach to facilitate comparisons of rates between different networks and units.

Full 2017 results for Necrotising enterocolitis and a description of the measure are found on page 124.

National Neonatal Audit Programme 2018 report on 2017 data

2.12. Minimising separation of mother and baby (term and late preterm)

Some babies admitted to neonatal units may be separated from their mothers for longer than necessary. It may be possible to care for some babies in transitional care, a setting which takes an interdisciplinary approach of both midwives and neonatal staff to deliver high-quality care to both mothers and babies and avoid their separation.11 This measure seeks to describe the number of babies admitted to neonatal units for low dependency care and to compare the number of days that babies were separated from their mothers.

The measure describes the number of "separation days" for each admission to a neonatal unit. Separation days are defined as days of low dependency care where breathing support was not needed. For some babies, separation from their mother may be able to be avoided altogether, with all their neonatal care delivered in a transitional care setting. For other babies where a neonatal unit admission is unavoidable, there may still be opportunities to reduce separation care days during admission, particularly where separation days are high.

Key findings

• Term babies admitted to neonatal units were separated from their mothers on average for 3.2 days on days when they received either “special care” without oxygen, or “normal care”. Networks varied in the average number of normal and special care days resulting in separation, and there is even greater variation between individual units (range 1 to 6 days) (Table 5.14.1, page 129, Table 5.14.2 page 130 and NNAP Online).

National Neonatal Audit Programme 2018 report on 2017 data

• The average number of term baby separation days is not higher for units providing higher levels of care (mean separation days at Special care units - 3.3 days; Local neonatal units - 3.2; Neonatal intensive care units - 3.1) (Table 5.14.1, page 129).

• Late preterm babies admitted to neonatal units had an average of 6.8 separation days (“special care” days where oxygen was not given, or “normal care” days) (Table 5.15.1, page 132).

• Networks vary in the average number of late preterm separation days (range 5.1 to 8 days), and there is even greater variation between individual units (range 1.3 to 12.5 days) (Table 5.15.2, page 133).

• The difference between units’ average late preterm separation days is not all explained by unit level, although variation does exist on average by unit level (mean separation days at special care units – 7.6 days; local neonatal units – 6.9 days; neonatal intensive care units – 6.3 days) (Table 5.15.1, page 132).

• Data completeness is good in the first year of this new measure as the relevant data are already routinely captured as part of care delivery.

• These findings suggest that opportunities exist to reduce separation of mothers and term and late preterm babies by providing some neonatal care as transitional care. Quality improvement activities based on this measure would be facilitated by presentation of the number of births at term in each hospital, data which is not yet available for England and Wales.

Recommendations

41. Neonatal units and trusts/health boards where transitional care cannot be delivered should work with their commissioners to develop the ability to deliver such care to minimise mother and baby separation, following the BAPM guidance A Framework for Neonatal Transitional Care.11

42. Neonatal units with above average numbers of separation days for term, or late preterm babies should consider if revision of their admission or discharge criteria and processes could reduce the number of mother and baby separation days.

43. Neonatal units should implement the BAPM guidance on the management of neonatal hypoglycaemia in term babies unless local circumstances make this inappropriate. Hypoglycaemia is a leading cause of term admission; some admissions for the management of hypoglycaemia could be avoided with the use of BAPM guidance.12

National Neonatal Audit Programme 2018 report on 2017 data

44. Neonatal units should be aware of their rates of admission for term babies, and use the themes emerging from ATAIN project reviews in England of term admissions to inform possible targeted review of their admission and discharge processes.

45. Neonatal networks should work collaboratively with local maternity system and maternity and neonatal safety collaborative colleagues (or their equivalents in Scotland and Wales) to understand the themes emerging from the ATAIN project and to assist their units in reducing unnecessary separation of the mother and her term baby.

46. The NNAP should seek to present the number of admissions and separation days alongside the number of births in each gestational age category.

Full 2017 results for Minimising inappropriate separation of mother and baby and a description of the measure are found on pages 127 and 131.

National Neonatal Audit Programme 2018 report on 2017 data

2.13. Breastmilk feeding at discharge home

Premature babies are vulnerable to infection, and their own mother’s milk provides an important line of defence through the protective antibodies that it provides. These significant health benefits include a reduction in infection and bowel problems, as well as improved longer-term health and neurodevelopmental outcomes.

Key findings

• Only half (6,418, 57%) of the 11,282 babies born at less than 33 weeks gestational age cared for in NNAP units were analysed for this audit measure. Most excluded babies were transferred away from their unit of birth.

• Six out of 10 (60.5%) babies were getting some breast milk at discharge but the rate of breastmilk feeding at discharge has improved only marginally over time (2016 – 59%, 2015 – 58%) (Table 5.16.3, page 137).

• At unit level there is marked variation in breast milk feeding rates at discharge for the most preterm babies, with NICUs ranging from 32.6% to 89.7% (NNAP Online).

• There is also marked network variation (48.6% to 87.8%), with networks in the north and west of the country generally having lower rates of breast milk feeding at discharge (Table 5.16.2, page 136). This is broadly in keeping with feeding practices for more mature babies not admitted to neonatal units.13

National Neonatal Audit Programme 2018 report on 2017 data

Recommendations

47. Neonatal units should use these data, alongside available data concerning breastfeeding practices in non-admitted babies in their local area, to inform local quality improvement activity aiming to improve rates of breastmilk feeding. Neonatal units can use The Baby Friendly Initiative (UNICEF)14 and the Bliss Baby Charter8 to support this activity.

48. The NNAP should develop a measure of early breastmilk feeding.

Full 2017 results for Breastmilk feeding at discharge home and a description of the measure are found on page 134.

National Neonatal Audit Programme 2018 report on 2017 data

2.14. Follow-up at two years of age

It is important that the development of very preterm babies is monitored by a paediatrician or neonatologist after the baby is discharged from the neonatal unit. This measure looks at whether there is a documented follow up consultation at two years of age for babies born at less than 30 weeks gestational age between July 2014 and June 2015 who survived and were discharged home from the neonatal unit. The follow up consultation assesses whether there are any significant problems with movement, the senses, and whether there are delays in development or other health problems. Babies born very early encounter these problems more often than those born at full term and it is important for those involved in the care of babies to know how the babies are developing as they get older so that they can arrange appropriate treatment.

Key findings

• Almost two in five (37.4%) of 4,043 babies born at less than 30 weeks gestational age between July 2014 and June 2015 were not recorded as having been seen for a follow up assessment at two years of age, despite the service specification and longstanding concern about practice in this area (Table 5.17.1, page 140).

• The proportion of babies documented as having been followed up at two years has increased only marginally since 2015 (2013 – 44.8%; 2014 – 54.5%; 2015 – 60%; 2016 – 60.9%; 2017 – 62.6%) (Table 5.17.3, page 141).

• Some units achieved good levels of follow up: 23 out of 45 units with more than 30 babies to follow up saw 70% or more of their babies. No network saw more than

National Neonatal Audit Programme 2018 report on 2017 data

three quarters of their babies, but one network saw fewer than half its babies (NNAP Online and Table 5.17.2, page 140).

• Six units achieved follow up of 100% of babies attributed to their units – three of these units had more than 10 babies to follow up. Forty-two units achieved follow up of 80% or more of the babies attributed to their units. Of these 35 had more than 10 babies to follow up (NNAP Online).

Recommendations

49. Neonatal units and networks should adopt the NICE guideline on Developmental follow-up of children and young people born preterm,33 and make progress with the implementation of pathways across organisational structures (e.g. Sustainability and transformation plan footprints in England). This requires a multidisciplinary, whole health economy approach.

50. Neonatal units with incomplete data capture should ensure that they have the processes in place to document follow up at two years of age.

51. Neonatal units should discuss arrangements for two-year follow up with families prior to discharge home of their baby, supported by written communication which includes the expected timeframe for the follow up consultation.

Full 2017 results for Follow-up at two years of age and a description of the measure are found on page 138.

National Neonatal Audit Programme 2018 report on 2017 data

3. Methods

3.1. Audit measures and measure development

The NNAP is responsive to changes in guidance and standards and to the needs of its stakeholders. It has an established process for developing new audit measures and reviewing and revising existing measures where necessary.

New measures are developed in the NNAP in response to a need identified by audit users, attendees at the annual NNAP/NDAU Collaborators Meeting, professional organisations, parent support organisations, neonatal networks, national initiatives or members of the NNAP Methodology and Dataset Group and Project Board.

As part of the development process, the proposal is reviewed to ensure that it aligns with existing activities and initiatives in neonatology, and with any relevant standards or guidelines. The measure is then developed by the NNAP Project Team and Methodology and Dataset Group in consultation with key stakeholders, before a final review and approval by the NNAP Project Board. Participating units are notified of any new or amended measures ahead of the start of the data collection period in which they will be used for the first time.

In July 2017, the Methodology and Dataset Group conducted a review of measures against current standards and guidelines and against potential for quality improvement. After this review, some measures were clarified or adjusted in line with the current evidence base, before inclusion in the set of measures for 2018 data. A review of measures to be included in the audit for the 2019 data year is currently underway.

National Neonatal Audit Programme 2018 report on 2017 data

3.2. Case ascertainment

Data for the NNAP analyses are extracted from the National Neonatal Research Database (NNRD) held at the Neonatal Data Analysis Unit (NDAU). The NNRD contains a predefined set of variables (the National Neonatal Dataset) obtained from the electronic neonatal patient records of each participating NHS Trust. Data are downloaded from the BadgerNet patient record system used in neonatal units and transferred to NDAU with health board and trust Caldicott Guardian approval. For Scotland a separate approval was received from the Public Benefit and Privacy Panel for health and social care.

In usual practice, every baby admitted to a participating neonatal unit entered on the BadgerNet patient record system, and is eligible for inclusion in NNAP; the audit therefore achieves 100% case ascertainment in the participating organisations. Babies receiving special care in transitional care areas or postnatal wards can also be entered, but it is known that some units do not enter data for such babies and for this reason measures do not concentrate on care outside neonatal units.

For most audit measures in this report, the cohort comprises all babies with a final discharge from neonatal care from 1 January to 31 December 2017. There are some exceptions to this; Encephalopathy and Minimising inappropriate separation of mother and baby (term and late preterm) measures use birth year, and Follow-up at two years of age comprises babies born between July 2014 and June 2015.

3.3. Data quality and completeness

For the 2017 data, quarterly reports were produced by the NNAP project team and disseminated to all neonatal unit NNAP clinical leads to provide regular updates on their data completeness and adherence to the NNAP standards. All neonatal units were provided with a summary report of their 2017 data in January 2018 after which they were given a final six-week window of opportunity to review and amend their 2017 data on the BadgerNet system. The final 2017 data download for this report was extracted from BadgerNet after the reviewing process had closed at the end of March 2018.

In the 2017 data report, we report outlier analysis and main report measures using a “no imputation” approach. By this we mean that rates of adherence to standards, or rates of clinical outcomes are described for the babies where the outcome is known. Numbers of patients with an outcome are included under “with outcome” in results tables. Missing data are also presented in results tables alongside clinical data.

National Neonatal Audit Programme 2018 report on 2017 data

3.4. Babies included in the 2017 dataset

The 2017 download included data on care provided for 104,183 babies with final discharge from neonatal care in 2017. The number of babies eligible for each audit question varies depending on the gestational age covered by the audit measure and the episode of care under consideration.

In addition, numerators may differ from the figures extracted locally; for example, in the analysis of the audit measure Consultation with parents some babies born, first admitted and discharged in 2017 do not appear in the analysis if they had a subsequent episode that continued into 2018.

Similarly, babies with episodes spanning years 2016 and 2017 were included in the data for year 2017. The NDAU conducts NNAP analyses using the age of the baby in minutes from birth, as opposed to calendar days to enhance patient anonymity. This can result in minor variations in the numerators for age critical fields, such as the timing of ROP screening.

A data cleaning and validation process is applied to the raw dataset before creation of the NNAP dataset that is used to produce the data included in this report. That process includes:

• Checking the providers included in the dataset against a master list to identify new providers.

• Removal of episodes which are complete duplicates or do not have birth year and gestation at birth or admission times entered.

• Only babies who were finally discharged in the NNAP reporting year of interest are kept in the NNAP dataset. The exception to this dataset is the cohort used for the Encephalopathy measure, this dataset is based on those babies with a birth year in the NNAP reporting year of interest.

National Neonatal Audit Programme 2018 report on 2017 data

3.5. Denominator data: live births

To support the analysis and interpretation of some of the audit measures, the NNAP has sought to use data on the number of live births at each hospital in England, Wales and Scotland. These data are not available at hospital level in England and Wales, and are not yet available at trust level for the calendar year 2017. Although these data are available for Scotland, we use 2016 data aggregated to trust or health board as a denominator for the Encephalopathy measure. These data from MBRRACE-UK for England and Wales, and from the Scottish Birth Record (managed by the Information Services Division) for Scotland. Next year, live birth data will be used to support the interpretation of the Minimising separation of mother and baby measures.

3.6. Outlier methodology and identification process

The NNAP selects measures for outlier analysis when data collection is mature and where identifying variation is likely to be useful for quality improvement purposes. Benchmarking is typically applied to new measures and where is a lack of a nationally agreed guideline or standard.

The NNAP manages outlier status in line with the RCPCH policy for the Detection and Management of Outlier Status for Clinical Indicators. 15 The RCPCH has agreed analytical models for identifying outliers as part of the statistical and analytical plan for each audit, but as a general principle the College bases the actions regarding outliers upon the HQIP 2017 guidance for management of outliers, Detection and management of outliers for National Clinical Audits. 16

For more information about the methodology used for the detection of outliers, refer to the NNAP Statistical Analysis Plan for 2017 data: www.rcpch.ac.uk/nnap.

National Neonatal Audit Programme 2018 report on 2017 data

3.7. Longitudinal analysis

This year, for the first time, NNAP is assessing the difference in performance on a range of audit measures over time. The change between two years is measured by the difference of the rates of compliance. Outliers among these changes, either exceptional reductions or improvements, are identified by the established methods and can be identified on NNAP Online. In the audit of 2017 data, comparisons are made for pairs of years 2016-2017 and 2015-2017.

For more information about the methodology used for longitudinal outlier analysis, refer to the NNAP Statistical Analysis Plan for 2017 data: www.rcpch.ac.uk/nnap.

3.8. Risk adjustment for bronchopulmonary dysplasia (BPD) or death

Risk adjustment for the combined outcome of BPD or death was conducted by comparing the prevalence of BPD or death in each unit and network with the prevalence of BPD in a set of babies with very similar background characteristics that were treated in the whole domain of the NNAP. The matching was conducted for an extensive set of background variables extracted from the NNRD.

The output of the matching method is a “treatment effect”. Treatment effect is the difference between the rate of BPD or death in babies cared for in a neonatal network compared to the observed rate for a matched group of babies with very similar case mix, cared for in all neonatal units. A positive treatment effect indicates that the rate of BPD or death is higher in the network of interest than for a comparable group of babies cared for in all neonatal units.

For more information about the methodology used for the matching method of BPD analysis, refer to Appendix E: Matching method of comparing outcomes for BPD and the NNAP Statistical Analysis Plan for 2017 data: www.rcpch.ac.uk/nnap.

National Neonatal Audit Programme 2018 report on 2017 data

3.9. Mortality reporting: Mortality to discharge in very preterm babies

Beginning in 2017, NNAP has been collecting data on mortality in very preterm babies. We will report mortality in babies born at 24-31 weeks gestational age with the first full three year rolling period being 2017-2019. Units are asked to report deaths in the usual way, but in addition deaths occurring before discharge home in babies of less than 44 weeks gestationally corrected age should be reported via BadgerNet, even if they occur in a non NNAP unit, or occur in a palliative care environment. Mortality will be reported only by network of birth.

For the full definition of this measure, please refer to the NNAP 2018 measures guide: https://www.rcpch.ac.uk/work-we-do/quality-improvement-patient-safety/national- neonatal-audit-programme-nnap/about.

National Neonatal Audit Programme 2018 report on 2017 data

3.10. Neonatal unit designations

We use the Toolkit for High Quality Neonatal Services (Department of Health, 2009) to define different levels of neonatal unit as follows:

Special care units (SCUs) provide special care for their own local population. Depending on arrangements within their neonatal network, they may also provide some high dependency services. In addition, SCUs provide a stabilisation facility for babies who need to be transferred to a neonatal intensive care unit (NICU) for intensive or high dependency care, and they also receive transfers from other network units for continuing special care.

Local neonatal units (LNUs) provide neonatal care for their own catchment population, except for the sickest babies. They provide all categories of neonatal care, but they transfer babies who require complex or longer-term intensive care to a NICU, as they are not staffed to provide longer-term intensive care. Most babies over 27 weeks gestational age will usually receive their full care, including short periods of intensive care, within their LNU. Some networks have agreed variations on this policy, due to local requirements. Some LNUs provide high dependency care and short periods of intensive care for their network population. LNUs may receive transfers from other neonatal services in the network, if these fall within their agreed work pattern.

Neonatal intensive care units (NICUs) are sited alongside specialist obstetric and feto- maternal medicine services, and provide the whole range of medical neonatal care for their local population, along with additional care for babies and their families referred from the neonatal network. Many NICUs are co-located with neonatal surgery services and other specialised services. Medical staff in a NICU should have no clinical responsibilities outside the neonatal and maternity services.

National Neonatal Audit Programme 2018 report on 2017 data

3.11. Neonatal network designations

Neonatal networks are designed to support the delivery of high quality neonatal care in their region for all their population. They ensure that mothers and babies are treated in a hospital appropriate to their needs.

In this report, we present data by neonatal Operational Delivery Network (ODN) in England. The Wales Neonatal Network comprises all neonatal units in Wales. In Scotland, there are currently three Managed Clinical Networks (MCN); West of Scotland, South East and Tayside, and North of Scotland. In line with recommendations made in The Best Start: A Five-Year Forward Plan for Maternity and Neonatal Care in Scotland, 17 Scotland will shortly be moving to a single neonatal network; pending this, and in agreement with the three existing networks, the NNAP reports Scottish unit data as a single network.

Where there is a change in network configuration, unit name or unit level, the NNAP will apply the status as at the end of the data reporting year. For example, if the configuration of a network changes on 1 April 2018, 2018 data will be presented as per the network configuration on 31 December 2018.

National Neonatal Audit Programme 2018 report on 2017 data

4. Case studies: how NNAP supports local quality improvement

4.1. Think magnesium! A multidisciplinary approach to improve magnesium sulphate uptake

Presented by

Dr Sankara Narayanan (Consultant Neonatologist & NNAP lead)1, Dr Anastasia Katana (Locum Consultant Neonatologist)1, Ms Nanda Shetty (Consultant Obstetrician)2, Ms Marcellina Coker (Consultant Obstetrician)2

Neonatal1 & Obstetrics2 Department, Watford General Hospital, West Hertfordshire NHS Trust

Background

Research has shown that magnesium sulphate (MgSO4), given antenatally in threatened preterm labour, is neuroprotective and reduces cerebral palsy. Watford General Hospital was an outlier for this audit measure in the 2017 NNAP report (2016 data). In this case study,

we demonstrate how we have used NNAP antenatal MgSO4 data to guide our quality

improvement in MgSO4 uptake.

Our improvement plan

Our improvement journey started in 2016, when we became aware that our neonatal

service was an outlier for antenatal magnesium sulphate (MgSO4) administration with an uptake approximately 20%, which was far below the national average (43%).

Strategy for change:

Using the Institute of Healthcare Improvement (IHI) model we aimed to increase the

uptake of MgSO4 in eligible preterm deliveries. Improvement was defined as an increase

in the uptake of MgSO4 from 15% in 2016 to 40%, hence reaching the then national average.

National Neonatal Audit Programme 2018 report on 2017 data

Primary, secondary drivers were identified which informed change ideas. These changes were tested in iterative plan, do, study, act (PDSA) cycles.

What we did:

Figure 4.1.1: Driver Diagram

Figure 4.1.2: Compliance to Magnesium Sulphate Administration - Run chart

Target

National Neonatal Audit Programme 2018 report on 2017 data

Change package:

We have increased awareness amongst the neonatal, obstetric and midwifery teams by presenting run chart data (Figure 2) in monthly Perinatal Mortality and Morbidity meetings.

We commenced frequent bitesize MgSO4 awareness and engagement sessions with all the stakeholders. A new and simple guideline for MgSO4 administration was developed and implemented and we encouraged 1:1 midwifery care in labour. Safety “huddles”, twice daily board rounds, were introduced to identify all eligible women. Additionally, we involved the service users, by providing them with a parental information leaflet with the antenatal counselling pack and MgSO4 information were included in the antenatal counselling conversations and golden hour care checklist.

Challenges:

The main challenge was to change staff’s ethos and perceptions, bring down the “barriers” and move away from the culture “this is how we do it for years and it works for our patients”. We achieved this by enhancing communication and knowledge sharing through safety huddles, identifying clinical champions to support the team, presentations in joint multidisciplinary team forums and empowering ward level leadership. Encouraging parent involvement raised further awareness of this important neuroprotective treatment.

Outcomes

Table 4.1.1: Antenatal MgSO4 administration annual uptake

Level 2 Neonatal Unit

Year Eligible mothers MgSO4 given MgSO4 not given

2016 20 3 (15%) 17 (85%)

2017 18 10 (55%) 8 (44%)

Jan-May 2018 6 5 (83%) 1 (17 %)

According to the NNAP 2016 report only 15% of eligible women received antenatal MgSO4 at Watford General Hospital, compared to a national average of 43%. Table 1 and Figure 2 shows steady improvement from 2016 to 2018 with the implementation of change ideas. The 2017 data show that we met and surpassed our improvement target achieving a compliance of 55%, well above the national average. The 2018 NNAP data shows further improvement and a sustainable change. This sustainable improvement has a direct impact on the long-term neurodevelopmental outcomes and by extension to the quality of life of preterm babies born at less than 30 weeks of gestation.

National Neonatal Audit Programme 2018 report on 2017 data

Top tips for implementation

• Identify an appropriate maternity-neonatal forum to share NNAP MgSO4 data • Identify and engage stakeholders and frontline champions within maternity and neonates • Use live NNAP dashboard on BadgerNet to generate live run charts • Partner with parents for improvement

Acknowledgements

Justine Chung, Matron, Delivery Suite, Bhavani Sivakumar, BadgerNet Data Analyst.

National Neonatal Audit Programme 2018 report on 2017 data

4.2. Improving Parental consultation within 24 hours of admission

Presented by

Dr Natasha Maddock, Consultant Neonatologist, The Royal Oldham Hospital

Mrs Beverley Scholes, NICU Data Quality Clerk, The Royal Oldham Hospital

Background

Communicating to families about their baby’s progress is very important. When the services at The Royal Oldham Hospital were first developed into a level 3 NICU we made a big improvement increasing the percentage of parents spoken to from less than 50% to 86%; a level that was comparable with our peers and other level 3 units within the Network.

However, despite regular communication and reminders we had been unable to improve on the number of parents who were seen by a senior member of staff within 24 hours of admission. In 2014 we spoke to 84% and in 2015 86%. Whilst reviewing the 2016 data in February 2017 it became apparent that this had not improved despite there being a general improvement across the UK and I suspected we would be an outlier (which we were) so I put a plan in place to try and improve our documented communication.

Measures

We measured the percentage of parents spoken to for all patients that were admitted to the neonatal unit, not just those that fulfilled the NNAP criteria, and put them in a run chart.

National Neonatal Audit Programme 2018 report on 2017 data

Our improvement plan

In March 2017 I worked with our data quality clerk to devise a run chart that looked at the percentage of parents that were spoken to on a weekly basis. The clerk would check the BadgerNet data quality of all the babies admitted in the past week, and complete the spreadsheet. She would then print off the chart each week and display on the staff notice board.

For the first few months the results were mentioned at the communication meeting. Whilst this was officially anonymous, consultants could identify themselves as we attend the unit for a week.

My consultant colleagues agreed that we needed to improve and were happy to trial this. Once in place they requested that if a consultant spoke to 100% of parents that they would receive a gold star.

Outcomes

• Our 2017 results have improved to 94%. • My consultant colleagues were very enthusiastic about this and worked hard to make sure it was successful. In fact they became very competitive and the gold star was their idea. • Within a few months there was a definite attitude change with medical staff always including first consultation as part of the presentation of an admission. • Despite less frequent updates, this improvement appears to have been sustained, although there is still room for improvement, particularly for those patients who remain on the unit for a short period of time or for an intervention.

Top tips for implementation

• You need a dedicated member of staff who can check the data and update the graph weekly. • Consultants are amazingly competitive.

Acknowledgements

I thank the Clinical Lead at Arrowe Park, Dr Oliver Rackham, for sharing his project.

If you would like a copy of the spreadsheet used by The Royal Oldham Hospital to devise their run chart, please contact [email protected].

National Neonatal Audit Programme 2018 report on 2017 data

4.3. Data collection of antenatal steroids given

Presented by

Hazel Williams, NNEB, BadgerNet System Data Officer at Calderdale & Huddersfield Foundation Trust.

Background

• My main aim was to improve our rates of administration of antenatal steroids. • Looking at our NNAP data for 2016, our rate for administration had dropped to 81% (from 93% in 2015). I wanted to see what had caused this change.

Measures

I planned to measure the improvement by comparing our 2017 data.

Our improvement plan

Firstly, I questioned what changes could have had an impact on our practice. The main change in our trust was that our maternity system had moved from our old PAS system to Athena (run by K2), and the maternity team had gone paperless.

We requested and established an interface between Athena and BadgerNet to pull the maternity details through for the baby admission when Athena replaced our old system. This has been working well, but the Athena system did not have a set place to record when steroids had been given. Information was being missed as it was being documented in various places not always obvious to neonatal staff.

Once this was realised, working with the maternity electronic patient record (EPR) midwives, we added a separate tab in Athena to capture that steroids had been given, the courses and the dates and times.

The next issue was to re-educate staff to document in the designated tab. Documentation was reviewed on a weekly basis and staff were notified by email of any errors relating to missing details.

National Neonatal Audit Programme 2018 report on 2017 data

At the same time, we notified the obstetric team of babies who did not have steroids documented so they could review why.

Outcomes

Our expected NNAP results indicate that we are up to 94.5% antenatal steroid administration for 2017, which is an excellent achievement. Our next project is to develop a tab within Athena to record administration of magnesium sulphate neuro protection, again this will also be beneficial for the maternity team.

Top tips for implementation

• Meet with maternity staff to discuss how they work with their EPR systems • Keep in touch with the maternity team, things change for them too in practice and EPR. • Find a “named” person in your IT department to work with who understands what you want to achieve.

Acknowledgements

• K2 and Clevermed, who were very helpful in assisting us with interfacing and testing of the system. The IT department at Calderdale. • Emma Hardwidge and Carol Gregson, Maternity EPR Midwifes at Calderdale, who have assisted throughout.

National Neonatal Audit Programme 2018 report on 2017 data

5. Full NNAP results

In this section, we present the full results, attributed by unit level and by network, for each of the NNAP measures.

For each measure, we include:

• NNAP audit measure • Change to audit measure for the 2017 data year • NNAP standard and source of standard • Whether outlier analysis will be conducted • Inclusion criteria • Results

For individual unit-level results, please go to NNAP Online.

Data completeness

Data completeness within the NNAP is good. Missing data are negligible for the measures Temperature, Bronchopulmonary dysplasia and Breastmilk feeding at discharge home. For Antenatal steroids, missing data are small at 1.5%, but have increased from 1.1% in 2016. Missing data for first consultation has also improved in the past year, but remains non- trivial at 2.7% (2016 - 3.6%).

Average rates of missing data by unit are given in Appendix A: Data completeness and unit level of participating units. Thirty-six units (20.1%) have average rates of missing data above 10%, and five units (2.8%) have rates above 20%. Aggregate rates of missing data are dominated by missing data for the Magnesium sulphate measure (2017 – 7.7%; 2016 – 18.8%).

National Neonatal Audit Programme 2018 report on 2017 data

5.1. Antenatal steroids

The key findings and recommendations for this audit measure are found on page 24.

NNAP audit measure

Is a mother who delivers a baby between 24 and 34 weeks gestational age inclusive given at least one dose of antenatal steroids?

Change to the audit measure for the 2017 data year: None.

From the 2018 data year, in line with NICE guidance, the NNAP measure will be amended to include babies at 23 weeks gestational age and exclude babies at 34 weeks gestational age.18,19,20,21

As noted in the Key findings and recommendations section, the NNAP will work with NMPA to consider whether antenatal steroid administration might best be audited within the NMPA in order that the administration of antenatal steroids to women who do not go on to deliver preterm can ultimately form part of the audit, and because the NMPA may have higher visibility among maternity caregivers. For this to function optimally, data on antenatal steroid administration should be collected as part of routine maternity data.

NNAP standard

Eighty-five percent (85%) of mothers should receive at least one dose of antenatal steroids.

Source of standard: NNAP Project Board

Outlier analysis: Yes

National Neonatal Audit Programme 2018 report on 2017 data

Inclusion criteria

Mothers of babies who meet the following criteria will be included in the analysis:

• Admitted for neonatal care • Experienced their final neonatal discharge in the calendar year of analysis • Had care provided by an NNAP unit • Gestational age at birth between 24 and 34 weeks inclusive • Only data from the first known episode of care will be considered for analysis • For multiple births, only one baby will be included so that each mother is only counted once per delivery

Results

18,965 eligible mothers were identified from data submitted for 21,661 babies by 179 neonatal units in England, Scotland and Wales and 33 places of birth not allied with an NNAP participating unit. Records for 67 babies were excluded from analysis because their data lacked sufficient detail to identify their mother.

At least one dose of antenatal steroids was administered to 88.6% of eligible mothers and no dose was administered to 11.4% of these mothers.

If the mother delivered at home, in transit, in an unknown location or in a maternity unit not allied to a NNAP participating unit, these results are not shown in Table 5.1.1. These cases are included in the network level Table 5.1.2. under ‘Other’ ODN.

There is 1.5% missing data and only four hospitals have missing data of over 10%. The number of eligible mothers in these four hospitals ranged from 12 to 63.

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.1.1: Administration of antenatal steroids, by neonatal unit level.

Mothers with babies in neonatal units participating in NNAP; deliveries at the gestational age of between 24 and 34 weeks, inclusive.

Administration of antenatal Missing NNU With NNU Mothers steroids data (%) level outcome Yes (%) No (%)

SCU 37 1,736 1,690 1,489 (88.1%) 201 (11.9%) 46 (2.6%)

LNU 88 7,998 7,902 7,043 (89.1%) 859 (10.9%) 96 (1.2%)

NICU 54 8,912 8,782 7,873 (89.6%) 909 (10.4%) 130 (1.5%)

1,969 Total 179 18,646 18,374 16,405 (89.3%) 272 (1.5%) (10.7%)

Figure 5.1.1: Caterpillar plot of the rates of administration of antenatal steroids; neonatal units.

Rates of administration of antenatal steroids are presented by dots, and 95% confidence intervals by vertical bars. The units are presented in the ascending order of the rates. The units can be identified in NNAP Online.

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.1.2: Administration of antenatal steroids, by neonatal network.

Mothers with babies in neonatal units participating in NNAP; deliveries at the gestational age of between 24 and 34 weeks, inclusive.

Administration of antenatal With Missing Network Mothers steroids outcome data (%) Yes (%) No (%) East of England 1,664 1,639 1,463 (89.3%) 176 (10.7%) 25 (1.5%) Neonatal ODN Midlands South West Newborn Neonatal 845 818 711 (86.9%) 107 (13.1%) 27 (3.2%) ODN North Central & North East London 1,320 1,294 1,150 (88.9%) 144 (11.1%) 26 (2%) Neonatal ODN North West London 770 768 719 (93.6%) 49 (6.4%) 2 (0.3%) Neonatal ODN North West Neonatal 2,323 2,278 2,068 (90.8%) 210 (9.2%) 45 (1.9%) ODN Northern Neonatal 747 734 642 (87.5%) 92 (12.5%) 13 (1.7%) ODN Scotland 1,203 1,188 1,098 (92.4%) 90 (7.6%) 15 (1.2%) South East Coast 1,354 1,343 1,195 (89%) 148 (11%) 11 (0.8%) Neonatal ODN South London 1,053 1,021 871 (85.3%) 150 (14.7%) 32 (3%) Neonatal ODN South West Neonatal 1,189 1,174 1,020 (86.9%) 154 (13.1%) 15 (1.3%) ODN Staffordshire, Shropshire and Black 668 663 596 (89.9%) 67 (10.1%) 5 (0.7%) Country Neonatal ODN Thames Valley & 1,396 1,391 1,249 (89.8%) 142 (10.2%) 5 (0.4%) Wessex ODN Trent Perinatal & Central Newborn 1,517 1,489 1,320 (88.7%) 169 (11.3%) 28 (1.8%) Neonatal ODN Wales 760 748 688 (92%) 60 (8%) 12 (1.6%) Yorkshire & Humber 1,819 1,808 1,599 (88.4%) 209 (11.6%) 11 (0.6%) Neonatal ODN Isle of Man 18 18 16 (88.9%) 2 (11.1%) 0 (0%) Other 319 299 145 (48.5%) 154 (51.5%) 20 (6.3%) Total 18,965 18,673 16,550 (88.6%) 2,123 (11.4%) 292 (1.5%)

National Neonatal Audit Programme 2018 report on 2017 data

Figure 5.1.2: Caterpillar plot of the rates of administration of antenatal steroids; neonatal networks.

Rates of administration of antenatal steroids are presented by black dots and the 95% confidence intervals are indicated by vertical bars. The networks are presented in the ascending order of the rates.

Table 5.1.3: Administration of antenatal steroids, by NNAP reporting year (2008 to 2017).

Administration of antenatal NNAP With Missing data NNU Mothers steroids year outcome (%) Yes (%) No (%) 2008 129 9,066 6,391 5,744 (89.9%) 647 (10.1%) 2,675 (29.5%) 2009 167 16,031 14,861 11,228 (75.6%) 3,633 (24.4%) 1,170 (7.3%) 2010 173 17,199 16,577 12,911 (77.9%) 3,666 (22.1%) 622 (3.6%) 2011 164 15,716 15,201 12,009 (79%) 3,192 (21%) 515 (3.3%) 2012 173 16,538 16,193 13,285 (82%) 2,908 (18%) 345 (2.1%) 2013 176 16,992 16,776 14,142 (84.3%) 2,634 (15.7%) 216 (1.3%) 2014 173 17,170 17,025 14,517 (85.3%) 2,508 (14.7%) 145 (0.8%) 2015 179 18,687 18,550 15,910 (85.8%) 2,640 (14.2%) 137 (0.7%) 2016 181 18,947 18,735 16,317 (87.1%) 2,418 (12.9%) 212 (1.1%) 2017 179 18,965 18,673 16,550 (88.6%) 2,123 (11.4%) 292 (1.5%)

National Neonatal Audit Programme 2018 report on 2017 data

5.2. Antenatal magnesium sulphate

Giving magnesium sulphate to women who are at risk of delivering a preterm baby reduces by 32% the chance that their baby will develop cerebral palsy.22 The NICE quality standard Preterm Labour and Birth recommends that all women that may deliver their baby at less than 30 weeks gestational age are offered magnesium sulphate where possible.18

The key findings and recommendations for this audit measure are found on page 26.

NNAP audit measure

Is a mother who delivers a baby below 30 weeks gestational age given magnesium sulphate in the 24 hours prior to delivery?

Change to the audit measure for the 2017 data year: None.

As noted in the Key findings and recommendations section, the NNAP will work with NMPA to consider whether antenatal magnesium sulphate administration might best be audited within the NMPA, in part because NMPA findings may have higher visibility among maternity caregivers. Inclusion of data collection about magnesium administration in routine datasets is therefore a key priority.

NNAP standard

Benchmarking

Inclusion criteria

Mothers of babies who met the following criteria were included for analysis:

• Final neonatal discharge in the calendar year of analysis • Had care provided by an NNAP unit • Gestational age at birth less than 30 weeks • Only the first known episode of care will be considered for analysis • For multiple births, only one baby will be included so that each mother is only counted once per delivery.

National Neonatal Audit Programme 2018 report on 2017 data

Results

There were 4,276 eligible mothers identified from data submitted for 4,843 babies by 176 neonatal units in England, Scotland and Wales and 20 places of birth not allied with an NNAP participating unit. Records for 16 babies were excluded from analysis because their data lacked sufficient detail to identify their mother.

Magnesium sulphate was administered to 64.1% of eligible mothers and no dose was administered to the remaining 35.9%. 8.0% of cases had missing or unknown magnesium sulphate data.

Almost half of units had no missing data for this measure, but 24 units have more than 25% missing data, and 8 units had more than 50% missing data. Missing data has markedly decreased since 2016 (2016 17.4%; 2017 8%, 323 babies).

If the mother delivered at home, in transit, in an unknown location or in a maternity unit not allied to a NNAP participating unit, these results are not shown in Table 5.2.1. These cases are included in the network level Table 5.2.2 under ‘Other’ network.

Table 5.2.1: Administration of magnesium sulphate, by neonatal unit level.

Mothers with babies in neonatal units participating in NNAP; deliveries at the gestational age of less than 30 weeks.

Administration of With Missing NNU level NNU Mothers magnesium sulphate outcome data (%) Yes (%) No (%) SCU 34 160 135 51 (37.8%) 84 (62.2%) 25 (15.6%)

LNU 88 1317 1213 687 (56.6%) 526 (43.4%) 104 (7.9%)

NICU 54 2700 2506 1768 (70.6%) 738 (29.4%) 194 (7.2%)

Total 176 4177 3854 2,506 (65%) 1,348 (35%) 323 (7.7%)

National Neonatal Audit Programme 2018 report on 2017 data

Figure 5.2.1: Caterpillar plot of the rates of compliance for administration of magnesium sulphate; neonatal units.

Rates of administration of magnesium sulphate are presented by black dots and the 95% confidence intervals by vertical bars. The units are presented in the ascending order of the rates. The units can be identified in NNAP Online.

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.2.2: Administration of magnesium sulphate, by neonatal network.

Mothers with babies in neonatal units participating in NNAP; deliveries at the gestational age of less than 30 weeks.

Administration of With Missing Network Mothers magnesium sulphate outcome data (%) Yes (%) No (%) East of England 318 305 185 (60.7%) 120 (39.3%) 13 (4.1%) Neonatal ODN Midlands South West Newborn Neonatal 207 163 84 (51.5%) 79 (48.5%) 44 (21.3%) ODN North Central & North East London Neonatal 289 262 169 (64.5%) 93 (35.5%) 27 (9.3%) ODN North West London 181 158 103 (65.2%) 55 (34.8%) 23 (12.7%) Neonatal ODN North West Neonatal 535 490 345 (70.4%) 145 (29.6%) 45 (8.4%) ODN Northern Neonatal 145 116 74 (63.8%) 42 (36.2%) 29 (20%) ODN Scotland 234 220 142 (64.5%) 78 (35.3%) 14 (6.0%) South East Coast 295 284 186 (65.5%) 98 (34.5%) 11 (3.7%) Neonatal ODN South London 296 276 190 (68.8%) 86 (31.2%) 20 (6.8%) Neonatal ODN South West Neonatal 255 244 180 (73.8%) 64 (26.2%) 11 (4.3%) ODN Staffordshire, Shropshire and Black 161 151 74 (49%) 77 (51%) 10 (6.2%) Country Neonatal ODN Thames Valley & 366 354 279 (78.8%) 75 (21.2%) 12 (3.3%) Wessex ODN Trent Perinatal & Central Newborn 318 287 177 (61.7%) 110 (38.3%) 31 (9.7%) Neonatal ODN Wales 172 155 93 (60%) 62 (40%) 17 (9.9%) Yorkshire & Humber 411 395 229 (58%) 166 (42%) 16 (3.9%) Neonatal ODN Isle of Man 1 1 0 (0%) 1 (100%) 0 (0%) Other 92 74 12 (16.2%) 62 (83.8%) 18 (19.6%) Total 4276 3935 2,522 (64.1%) 1,413 (35.9%) 341 (8.0%)

National Neonatal Audit Programme 2018 report on 2017 data

Figure 5.2.2: Caterpillar plot of the rates of compliance for administration of magnesium sulphate; neonatal networks.

Rates of administration of magnesium sulphate. The estimates are marked by black dots and the 95% confidence intervals are indicated by vertical bars. The networks are presented in the ascending order of the rates in 2017.

Table 5.2.3: Administration of antenatal magnesium sulphate, by NNAP reporting year (2016–2017)*.

Administration of With Missing data NNAP Year NNU Mothers magnesium sulphate outcome (%) Yes (%) No (%) 2016 182 4,242 3,506 1,868 (53.3%) 1,638 (46.7%) 736 (17.4%) 2017 176 4,276 3,935 2,522 (64.1%) 1,413 (35.9%) 341 (8%)

*Results presented here for 2016 and 2017 are both calculated using the 2017 measure derivation method so that they are directly comparable.

National Neonatal Audit Programme 2018 report on 2017 data

5.3. Birth in a centre with a neonatal intensive care unit (NICU)

Babies who are born at less than 27 weeks gestational age are at high risk of death and serious illness. National recommendations in England state that neonatal networks should aim to configure and deliver services to increase the proportion of babies at this gestational age that are delivered in a hospital with a neonatal intensive care unit (NICU) on site.23 This is because there is evidence that outcomes are improved if such immature babies are cared for in a NICU from birth.

The key findings and recommendations for this audit measure are found on page 28.

NNAP audit measure

Is an admitted baby born at less than 27 weeks gestational age delivered in a maternity service on the same site as a designated NICU?24

Change to the audit measure for the 2017 data year: New measure for 2017 data year.

NNAP standard

Benchmarking

Outlier analysis: Yes, at network level only.

Inclusion criteria

Babies were included for analysis if they meet the following criteria:

• Final neonatal discharge in the calendar year of analysis • Had care provided by an NNAP unit • Gestational age at birth of less than 27 weeks • Only data from the first known episode of care will be considered for analysis • For multiple births, only one baby will be included so that each mother is only counted once per delivery.

National Neonatal Audit Programme 2018 report on 2017 data

Results

1,621 babies were born less than 27 weeks across units in England, Scotland and Wales. Of these 73.9% were delivered appropriately in a hospital with a NICU on site.

There was no obvious correlation between workload of network and the proportion of babies delivered in a centre with a NICU (Table 5.3.1).

Table 5.3.1: Delivery location of babies born at less than 27 weeks, by neonatal network

Delivery location Hospital with Network Babies designated NICU Other (%) (%) East of England Neonatal ODN 110 51 (46.4%) 59 (53.6%) Midlands South West Newborn 77 61 (79.2%) 16 (20.8%) Neonatal ODN North Central & North East London 125 87 (69.6%) 38 (30.4%) Neonatal ODN North West London Neonatal ODN 71 56 (78.9%) 15 (21.1%) North West Neonatal ODN 247 202 (81.8%) 45 (18.2%) Northern Neonatal ODN 56 50 (89.3%) 6 (10.7%) Scotland 78 70 (89.7%) 8 (10.3%) South East Coast Neonatal ODN 124 99 (79.8%) 25 (20.2%) South London Neonatal ODN 119 90 (75.6%) 29 (2.4%) South West Neonatal ODN 94 64 (68.1%) 30 (31.9%) Staffordshire, Shropshire and Black 66 45 (68.2%) 21 (31.8%) Country Neonatal ODN Thames Valley & Wessex ODN 137 104 (75.9%) 33 (24.1%) Trent Perinatal & Central Newborn 125 85 (68%) 40 (32%) Neonatal ODN Wales 53 39 (73.6%) 14 (26.4%) Yorkshire & Humber Neonatal ODN 138 95 (68.8%) 43 (31.2%) Isle of Man 1 0 (0%) 1 (100%) Total 1,621 1,198 (73.9%) 418.349 (26.1%)

National Neonatal Audit Programme 2018 report on 2017 data

Figure 5.3.1: Caterpillar plot of the rates of compliance for Birth in a centre with a NICU; neonatal networks.

National Neonatal Audit Programme 2018 report on 2017 data

5.4. Promoting normal temperature on admission for very preterm babies

The key findings and recommendations for this audit measure are found on page 30.

NNAP audit measure

Does an admitted baby born at less than 32 weeks gestational age have its first measured temperature of 36.5–37.5°C within one hour of birth?

Change to the audit measure for the 2017 data year: Clarification that the standard relates to achievement of normal first temperature within an hour of birth.

NNAP standard

The temperature should be taken for at least 98% of babies. The composite measure of timeliness and normal temperature should be met for at least 90% of babies. It is recognised that this is an aspirational standard.

Source of standard: NNAP Project Board

Outlier analysis: Yes

Inclusion criteria

Babies were included for analysis if they met the following criteria:

• Gestational age at birth of less than 32 weeks • Admitted to a neonatal unit within an hour of birth • Final neonatal discharge in the calendar year of analysis

National Neonatal Audit Programme 2018 report on 2017 data

• Had care provided by an NNAP unit • Only the first known episode of care will be considered for analysis

Results

8,019 babies were born at a gestational age of less than 32 weeks in 178 NNAP units and 10 places of birth not associated with an NNAP participating unit. For 22 babies (0.3%) temperature data was missing. Of the 7,997 babies with recorded outcomes, the temperature was taken on time and within a normal range for 5147 babies (64.4%), late for 239 (3.0%), and was not taken in 15 cases (0.2%).

In the following tables responses are assigned “Other” if the mother delivered at home, in transit, in an unknown location or in a non-NNAP unit.

LNUs and NICUs achieve similar levels of thermoregulation (SCUs 59.9%; LNUs 64.9%; NICUs 65.1%).

Table 5.4.1: Timeliness and normothermia, by neonatal unit level.

Temperature taken within one hour of birth (on time and within a normal range).

Temperature measurement Missing NNU With On time (%) NNU Babies Not data level outcome Late < 32°C- 36°C- 36.5°C- > taken (%) 32°C 35.9°C 36.4°C 37.5°C 37.5°C 26 81 249 SCU 36 419 416 0 42 17 1 3 (0.7%) (6.3%) (19.5%) (59.9%) 168 457 1,854 LNU 88 2,863 2,857 0 293 76 9 6 (0.2%) (5.9%) (16%) (64.9%) 215 619 3,023 12 NICU 54 4,657 4,645 1 643 139 5 (4.6%) (13.3%) (65.1%) (0.3%) 35 14 21 Other - 80 79 1 1 7 0 1 (1.3%) (44.3%) (17.7%) (26.6%) 444 1,171 5,147 979 239 15 22 Total 178 8,019 7,997 2 (5.6%) (14.6%) (64.4%) (12.2%) (3.0%) (0.2%) (0.3%)

Figure 5.4.1: Caterpillar plot of the rates of compliance for temperature on admission; neonatal units.

Rates of compliance with the standard for temperature on admission (on time and within a normal range). The estimated rates of compliance with the standard are marked by black

National Neonatal Audit Programme 2018 report on 2017 data dots and the 95% confidence intervals by vertical bars. The units are presented in the ascending order of the rates. The units can be identified in NNAP Online.

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.4.2: Timeliness and normothermia, by neonatal network

Temperature taken within one hour of birth (on time and within a normal range).

Temperature measurement Missing With On time (%) Network Babies Not data outcome < 32°C- 36°C- 36.5°C-37.5°C Late >37.5°C taken (%) 32°C 35.9°C 36.4°C (normothermia) 84 48 East of England Neonatal ODN 615 613 0 31 425 (69.3%) 23 2 2 (0.3%) (13.7%) (7.8%) Midlands South West Newborn 57 56 357 356 0 38 180 (50.6%) 25 0 1 (0.3%) Neonatal ODN (16%) (15.7%) North Central & North East London 111 37 554 552 0 50 340 (61.6%) 14 0 2 (0.4%) Neonatal ODN (20.1%) (6.7%) 38 31 2 North West London Neonatal ODN 341 339 0 10 250 (73.7%) 10 0 (11.2%) (9.1%) (0.6%) 178 144 North West Neonatal ODN 1,009 1,006 0 51 604 (60%) 25 4 3 (0.3%) (17.7%) (14.3%) 46 41 Northern Neonatal ODN 305 303 0 16 198 (65.3%) 2 0 2 (0.7%) (15.2%) (13.5%) 52 57 Scotland 460 459 0 26 310 (67.5%) 14 0 1 (0.2%) (11.3%) (12.4%) 86 83 South East Coast Neonatal ODN 576 575 1 36 363 (63.1%) 6 0 1 (0.2%) (15%) (14.4%) 77 72 South London Neonatal ODN 496 491 0 27 302 (61.5%) 11 2 5 (1%) (15.7%) (14.7%) 0 78 South West Neonatal ODN 505 505 0 26 306 (60.6%) 33 2 0 (0%) (11.9%) (15.4%)

National Neonatal Audit Programme 2018 report on 2017 data

Staffordshire, Shropshire and Black 49 23 279 279 0 19 185 (66.3%) 3 0 0 (0%) Country Neonatal ODN (17.6%) (8.2%) 53 109 Thames Valley & Wessex ODN 688 688 0 11 510 (74.1%) 4 1 0 (0%) (7.7%) (15.8%) Trent Perinatal & Central Newborn 122 56 637 636 0 24 392 (61.6%) 40 2 1 (0.2%) Neonatal ODN (19.2%) (8.8%) 34 32 Wales 338 337 0 13 239 (70.9%) 18 1 1 (0.3%) (10.1%) (9.5%) 110 111 Yorkshire & Humber Neonatal ODN 776 776 0 31 519 (66.9%) 4 1 0 (0%) (14.2%) (14.3%) Isle of Man 3 3 0 0 0 (0%) 3 (100%) 0 (0%) 0 0 0 (0%) 14 Other 80 79 1 35 21 (26.6%) 1 (1.3%) 7 0 1 (1.3%) (17.7%) 1,171 979 22 Total 8,019 7,997 2 444 5,147 (64.4%) 239 15 (14.6%) (12.2%) (0.3%)

National Neonatal Audit Programme 2018 report on 2017 data

Figure 5.4.2: Caterpillar plot of the rates of compliance for temperature on admission; neonatal networks.

Rates of compliance with the standard for temperature on admission (on time and within a normal range). The estimated rates of compliance with the standard are marked by black dots and the 95% confidence intervals are indicated by vertical bars. The networks are presented in the ascending order of the rates in 2017.

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.4.3: Timeliness and normothermia, by NNAP reporting year (2013–2017).

Temperature measurement

NNAP With On time (%) Missing NNU Babies Not year outcome Late data (%) < 32°C- 36°C- 36.5°C-37.5°C taken >37.5°C 32°C 35.9°C 36.4°C (normothermia) 553 325 2013 170 2,908 2,848 2 334 1,485 (52.1%) 133 16 60 (2.1%) (19.4%) (11.4%)

611 2014 167 3,109 3,074 4 361 1,578 (51.3%) 380 131 9 35 (1.1%) (19.9%) (12.6%) 1,403 760 2015* 177 7,864 7,807 3 648 4,537 (58.1%) 432 24 57 (0.7%) (18%) (9.7%) 1,368 2016* 181 8,044 8,006 3 559 4,868 (60.8%) 960 (12%) 235 13 38 (0.5%) (17.1%) 1,171 979 2017* 178 8,019 7,997 2 444 5,147 (64.4%) 239 15 22 (0.3%) (14.6%) (12.2%)

*For 2015-2017 data babies born less than 32 weeks were included in the audit measure. In previous years, only babies less than 29 weeks were included.

National Neonatal Audit Programme 2018 report on 2017 data

5.5. Parental consultation within 24 hours of admission

The key findings and recommendations for this audit measure are found on page 32.

NNAP audit measure

Is there a documented consultation with parents by a senior member of the neonatal team within 24 hours of a baby’s first admission?25,26,27

Note: By senior member of the neonatal team, NNAP means a consultant or middle grade doctor, or a nurse practitioner acting in such a role.

Change to the audit measure for the 2017 data year: None.

Outlier analysis: Yes

NNAP standard

A consultation should take place within 24 hours of first admission for every baby.

Source of standard: NNAP Project Board

Inclusion criteria Babies were included for analysis if they met the following criteria:

• Final neonatal discharge in the calendar year of analysis • Had care provided by an NNAP unit • Admitted to neonatal care for at least 12 hours in their first episode, receiving special care or a higher level of neonatal care • Babies receiving neonatal care in non-neonatal unit locations (postnatal ward, transitional care etc.) were not included.

National Neonatal Audit Programme 2018 report on 2017 data

Results

There were 93,859 first episodes of care (lasting at least 12 hours) reported by 179 neonatal units considered for this question. Babies who did not receive HRG 1, 2 or 3 on a neonatal unit during their first day of care or whose admission was for less than 12 hours were excluded from the analysis; this left 59,655 first episodes eligible for the audit measure.

A senior member of the neonatal team consulted parents or carers within 24 hours of admission for 94.6% of eligible episodes. Consultations that occurred before admission or more than 24 hours after admission were recorded in 3.3% of eligible episodes.

No consultation occurred for 2.2% of eligible episodes and data on consultations was either missing or ‘unknown’ for 2.7% of eligible episodes.

Table 5.5.1: Time of first consultation, by neonatal unit level.

First consultation with a senior member of staff within 24 hours of first admission

Time of first consultation Within 24 NNU With Missing NNU Babies hours of Before After 24 No level outcome data (%) admission admission hours consultation (%) 5,450 266 SCU 37 6,168 5,902 184 106 162 (92.3%) (4.3%) 24,543 686 LNU 88 26,413 25,727 425 334 425 (95.4%) (2.6%) 24,898 673 NICU 54 27,074 26,401 350 487 666 (94.3%) (2.5%) 54,891 927 1625 Total 179 59,655 58,030 959 (1.7%) 1,253 (2.2%) (94.6%) (1.6%) (2.7%)

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.5.2: Time of first consultation, by neonatal network.

First consultation with a senior member of staff within 24 hours of first admission.

Time of first consultation With Missing Network Babies Within 24 hours Before After 24 No outcome data (%) of admission (%) admission hours consultation

East of England Neonatal ODN 6,637 6,389 6,003 (94%) 186 64 136 248 (3.7%) Midlands South West Newborn Neonatal ODN 2,807 2,730 2,486 (91.1%) 71 77 96 77 (2.7%) North Central & North East London Neonatal ODN 4,593 4,512 4,403 (97.6%) 36 45 28 81 (1.8%) North West London Neonatal ODN 2,374 2,278 2,138 (93.9%) 39 56 45 96 (4%) North West Neonatal ODN 7,060 6,914 6,493 (93.9%) 62 161 198 146 (2.1%) Northern Neonatal ODN 2,155 2,108 2,010 (95.4%) 38 19 41 47 (2.2%) Scotland 4,239 4,125 3,828 (92.8%) 105 57 135 114 (2.7%) South East Coast Neonatal ODN 3,772 3,684 3,566 (96.8%) 34 44 40 88 (2.3%) South London Neonatal ODN 3,723 3,539 3,384 (95.6%) 50 46 59 184 (4.9%) South West Neonatal ODN 4,057 3,879 3,490 (90%) 103 144 142 178 (4.4%) Staffordshire, Shropshire and Black Country 1,905 1,895 1,762 (93%) 30 32 71 10 (0.5%) Neonatal ODN Thames Valley & Wessex ODN 4,666 4,646 4,616 (99.4%) 4 11 15 20 (0.4%) Trent Perinatal & Central Newborn Neonatal ODN 4,421 4,258 3,983 (93.5%) 72 88 115 163 (3.7%) Wales 2,416 2,324 2,176 (93.6%) 74 25 49 92 (3.8%) Yorkshire & Humber Neonatal ODN 4,730 4,652 4,460 (95.9%) 52 57 83 78 (1.6%) Isle of Man 100 97 93 (95.9%) 3 1 0 3 (3%)

Total 59,655 58,030 54,891 (94.6%) 959 (1.7%) 927 (1.6%) 1,253 (2.2%) 1625 (2.7%)

National Neonatal Audit Programme 2018 report on 2017 data

Figure 5.5.1: Caterpillar plot of the rates of compliance for first consultation within 24 hours of admission; neonatal networks.

Rates of compliance with the standard for first consultation within 24 hours of admission. The estimated rates of compliance with the standard are marked by black dots and the 95% confidence intervals are indicated by vertical bars. The networks are presented in the ascending order of the rates.

Table 5.5.3: Time of first consultation, by NNAP reporting year (2013–2017).

Time of first consultation NNAP First With Missing NNU Within 24 After 24 Before No Year episodes outcome data (%) hours (%) hours admission consultation 42,807 2,736 2013 176 50,757 48,021 1,386 2,273 1,555 (89.1%) (5.4%) 46,485 1704 2014 174 52,372 50,668 1,451 1,134 1,598 (91.7%) (3.3%) 51,300 2237 2015 179 58,077 55,840 1,261 1,204 2,075 (91.9%) (3.9%) 54,422 2222 2016 181 60,148 57,926 1,054 1,024 1,426 (94%) (3.7%) 54,891 1625 2017 179 59,655 58,030 927 959 1,253 (94.6%) (2.7%)

National Neonatal Audit Programme 2018 report on 2017 data

5.6. Parental presence at consultant ward rounds

The key findings and recommendations for this audit measure are found on page 34.

NNAP audit measure

For a baby admitted for more than 24 hours, did at least one parent attend a consultant ward round at any point during the baby’s admission?25,26,28

Note: Consultant ward round refers to any ward round where a consultant is in attendance, reviewing the care of patients, at any time of the day.

Change to the audit measure for the 2017 data year: New measure for the 2017 data year.

The NNAP Project Board and Methodology and Dataset Group recommended that the NNAP should clarify that this measure is designed to address the inclusion of parents on a ward round where decisions are made rather than updating parents at another time.

NNAP standard

Benchmarking.

National Neonatal Audit Programme 2018 report on 2017 data

Inclusion criteria

Babies were included for analysis if they met the following criteria:

• Experienced their final neonatal discharge in the calendar year of analysis • Had care provided by an NNAP unit • At least 24 hours (≥ 1440 minutes) between the admission time and discharge time for the episode of care • A baby may have several admissions; every eligible admission is included in the analysis • Babies receiving all neonatal care in non-neonatal unit locations (postnatal ward, transitional care etc.) are not included.

Results

There were 71,622 admissions for babies admitted for more than 24 hours. Of these admissions, 18.8% had missing data, leaving 58,163 eligible admissions. 74.3% of these eligible admissions had the presence of at least one parent for one or more ward rounds recorded at any point during the admission.

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.6.1: Parent present on one or more consultant ward rounds, by neonatal unit level.

Length of Parental presence on consultant ward round Missing data (%) NNU With NNU stay Admissions Parent not level outcome Parent present (%) No ward round (days) present ≤ 7 4,621 3,470 2,332 (67.2%) 352 786 1,151 (24.9%) 8-14 1,499 1,329 1,050 (79%) 102 177 170 (11.3%) 15-21 836 769 628 (81.7%) 64 77 67 (8%) SCU 37 22-28 461 419 376 (89.7%) 25 18 42 (9.1%) >28 566 520 439 (84.4%) 44 37 46 (8.1%) Total 7,983 6,507 4,825 (74.2%) 587 1,095 1,476 (18.5%) ≤ 7 16,916 13,132 9,598 (73.1%) 2,137 1,397 3,784 (22.4%) 8-14 5,685 5,154 4,206 (81.6%) 564 384 531 (9.3%) 15-21 2,987 2,829 2,259 (79.9%) 322 248 158 (5.3%) LNU 88 22-28 1,592 1,519 1,268 (83.5%) 142 109 73 (4.6%) >28 3,672 3,569 3,139 (88%) 218 212 103 (2.8%) Total 30,852 26,203 20,470 (78.1%) 3,383 2,350 4,649 (15.1%) ≤ 7 16,836 10,866 6,532 (60.1%) 2,990 1,344 5,970 (35.5%) 8-14 6,193 5,363 3,856 (71.9%) 917 590 830 (13.4%) 15-21 3,083 2,835 2,129 (75.1%) 404 302 248 (8%) NICU 54 22-28 1,746 1,652 1,273 (77.1%) 214 165 94 (5.4%) >28 4,929 4,737 4,145 (87.5%) 277 315 192 (3.9%) Total 32,787 25,453 17,935 (70.5%) 4,802 2,716 7334 (22.4%) Total 179 Total 71,622 58,163 43,230 (74.3%) 8,772 (15.1%) 6,161 (10.6%) 13,459 (18.8%)

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.6.2: Parent present on one or more consultant ward rounds, by neonatal network.

Parental presence on consultant ward Length of stay With round Missing Network Admissions (days) outcome Parent Parent not No ward data (%) present (%) present round ≤ 7 4,910 3812 2,802 (73.5%) 565 445 1,098 (22.4%) 8-14 1,459 1356 1,074 (79.2%) 193 89 103 (7.1%) 15-21 617 607 483 (79.6%) 74 50 10 (1.6%) East of England Neonatal ODN 22-28 349 340 283 (83.2%) 34 23 9 (2.6%) >28 746 737 651 (88.3%) 45 41 9 (1.2%) Total 8,081 6852 5,293 (77.2%) 911 648 1,229 (15.2%) ≤ 7 2,028 908 618 (68.1%) 192 98 1,120 (55.2%) 8-14 640 465 315 (67.7%) 112 38 175 (27.3%) Midlands South West Newborn 15-21 288 239 155 (64.9%) 71 13 49 (17%) Neonatal ODN 22-28 176 162 118 (72.8%) 38 6 14 (8%) >28 411 393 302 (76.8%) 69 22 18 (4.4%) Total 3,543 2167 1,508 (69.6%) 482 177 1,376 (38.8%) ≤ 7 3,097 2001 1,641 (82%) 293 67 1,096 (35.4%) 8-14 940 776 678 (87.4%) 88 10 164 (17.4%) North Central & North East London 15-21 511 430 367 (85.3%) 56 7 81 (15.9%) Neonatal ODN 22-28 292 237 205 (86.5%) 27 5 55 (18.8%) >28 781 664 583 (87.8%) 70 11 117 (15%) Total 5,621 4108 3,474 (84.6%) 534 100 1513 (26.9%) ≤ 7 1,463 602 420 (69.8%) 153 29 861 (58.9%) 8-14 500 319 232 (72.7%) 66 21 181 (36.2%) North West London Neonatal ODN 15-21 266 196 141 (71.9%) 47 8 70 (26.3%) 22-28 173 127 92 (72.4%) 28 7 46 (26.6%) >28 401 342 274 (80.1%) 54 14 59 (14.7%)

National Neonatal Audit Programme 2018 report on 2017 data

Total 2,803 1586 1,159 (73.1%) 348 79 1,217 (43.4%) ≤ 7 3,965 2718 1,507 (55.4%) 774 437 1,247 (31.5%) 8-14 1,694 1494 998 (66.8%) 288 208 200 (11.8%) North West Neonatal ODN 15-21 942 892 582 (65.2%) 187 123 50 (5.3%) 22-28 465 457 303 (66.3%) 81 73 8 (1.7%) >28 1,090 1082 881 (81.4%) 74 127 8 (0.7%) Total 8,156 6643 4,271 (64.3%) 1,404 968 1,513 (18.6%) ≤ 7 1,246 1025 707 (69%) 208 110 221 (17.7%) 8-14 598 544 430 (79%) 57 57 54 (9%) 15-21 290 267 225 (84.3%) 21 21 23 (7.9%) Northern Neonatal ODN 22-28 163 151 139 (92.1%) 6 6 12 (7.4%) >28 302 272 252 (92.6%) 10 10 30 (9.9%) Total 2,599 2259 1,753 (77.6%) 302 204 340 (13.1%) ≤ 7 2,386 1781 1,005 (56.4%) 519 257 605 (25.4%) 8-14 888 821 586 (71.4%) 111 124 67 (7.5%) 15-21 501 468 348 (74.4%) 39 81 33 (6.6%) Scotland 22-28 238 235 182 (77.4%) 22 31 3 (1.3%) >28 611 603 508 (84.2%) 24 71 8 (1.3%) Total 4,624 3908 2,629 (67.3%) 715 564 716 (15.5%) ≤ 7 2,710 2164 1,435 (66.3%) 388 341 546 (20.1%) 8-14 933 855 667 (78%) 126 62 78 (8.4%) 15-21 455 437 358 (81.9%) 54 25 18 (4%) South East Coast Neonatal ODN 22-28 246 242 210 (86.8%) 26 6 4 (1.6%) >28 591 584 518 (88.7%) 48 18 7 (1.2%) Total 4,935 4282 3,188 (74.5%) 642 452 653 (13.2%) ≤ 7 2,364 1280 865 (67.6%) 273 142 1,084 (45.9%) South London Neonatal ODN 8-14 838 607 406 (66.9%) 124 77 231 (27.6%)

National Neonatal Audit Programme 2018 report on 2017 data

15-21 459 385 264 (68.6%) 72 49 74 (16.1%) 22-28 235 196 146 (74.5%) 30 20 39 (16.6%) >28 599 544 461 (84.7%) 52 31 55 (9.2%) Total 4,495 3012 2,142 (71.1%) 551 319 1,483 (33%) ≤ 7 2,900 2162 1,686 (78%) 307 169 738 (25.4%) 8-14 911 839 747 (89%) 73 19 72 (7.9%) 15-21 453 438 414 (94.5%) 15 9 15 (3.3%) South West Neonatal ODN 22-28 268 262 251 (95.8%) 5 6 6 (2.2%) >28 590 584 571 (97.8%) 9 4 6 (1%) Total 5,122 4285 3,669 (85.6%) 409 207 837 (16.3%) ≤ 7 1,193 883 530 (60%) 179 174 310 (26%) 8-14 424 401 278 (69.3%) 61 62 23 (5.4%) Staffordshire, Shropshire and Black 15-21 213 209 141 (67.5%) 35 33 4 (1.9%) Country Neonatal ODN 22-28 100 99 76 (76.8%) 12 11 1 (1%) >28 316 315 285 (90.5%) 9 21 1 (0.3%) Total 2,246 1907 1,310 (68.7%) 296 301 339 (15.1%) ≤ 7 2885 2680 2,049 (76.5%) 415 216 205 (7.1%) 8-14 862 852 794 (93.2%) 29 29 10 (1.2%) 15-21 472 471 442 (93.8%) 17 12 1 (0.2%) Thames Valley & Wessex ODN 22-28 290 289 281 (97.2%) 3 5 1 (0.3%) >28 773 769 757 (98.4%) 9 3 4 (0.5%) Total 5,282 5061 4,323 (85.4%) 473 265 221 (4.2%) ≤ 7 2,971 2062 1,215 (58.9%) 471 376 909 (30.6%) 8-14 880 832 658 (79.1%) 104 70 48 (5.5%) Trent Perinatal & Central Newborn 15-21 491 483 370 (76.6%) 46 67 8 (1.6%) Neonatal ODN 22-28 276 273 219 (80.2%) 30 24 3 (1.1%) >28 659 653 581 (89%) 25 47 6 (0.9%)

100

National Neonatal Audit Programme 2018 report on 2017 data

Total 5,277 4303 3,043 (70.7%) 676 584 974 (18.5%) ≤ 7 1,453 1199 658 (54.9%) 320 221 254 (17.5%)

8-14 491 475 336 (70.7%) 69 70 16 (3.3%) Wales 15-21 278 269 209 (77.7%) 19 41 9 (3.2%) 22-28 178 176 137 (77.8%) 18 21 2 (1.1%) >28 409 405 348 (85.9%) 17 40 4 (1%)

Total 2,809 2524 1,688 (66.9%) 443 393 285 (10.1%) ≤ 7 2,747 2150 1,296 (60.3%) 413 441 597 (21.7%) 8-14 1,295 1188 901 (75.8%) 74 213 107 (8.3%) 15-21 653 626 508 (81.2%) 33 85 27 (4.1%) Yorkshire & Humber Neonatal ODN 22-28 349 343 274 (79.9%) 21 48 6 (1.7%) >28 878 870 743 (85.4%) 23 104 8 (0.9%) Total 5,922 5177 3,722 (71.9%) 564 891 745 (12.6%) ≤ 7 55 41 28 (68.3%) 9 4 14 (25.5%) 8-14 24 22 12 (54.5%) 8 2 2 (8.3%) 15-21 17 16 9 (56.3%) 4 3 1 (5.9%) Isle of Man 22-28 1 1 1 (100%) 0 0 0 (0%) >28 10 9 8 (88.9%) 1 0 1 (10%) Total 107 89 58 (65.2%) 22 9 18 (16.8%) 43,230 6,161 13,459 Total 71,622 58,163 8,772 (15.1%) (74.3%) (10.6%) (18.8%)

101

National Neonatal Audit Programme 2018 report on 2017 data

5.7. On-time screening for retinopathy of prematurity

The key findings and recommendations for this audit measure are found on page 36.

NNAP audit measure

Does an admitted baby born weighing less than 1501g, or at gestational age of less than 32 weeks, undergo the first retinopathy of prematurity (ROP) screening in accordance with the NNAP interpretation of the current guideline recommendations?29

Change to the audit measure for the 2017 data year: None.

NNAP standard

100% of eligible babies should receive ROP screening within the time windows for first screening recommended in the guidelines.

Note: In interpreting the national standards for this NNAP analysis, the Project Board has decided that a baby will be seen as having had ROP screening “on-time” if:

• A baby who was discharged before the ROP screening window opened had their first screening conducted prior to discharge, or • A ROP screen takes place within the ROP screening window, before or after discharge.

The NNAP Project Board has also agreed to allow an extra week either side of the ROP screening window.

102

National Neonatal Audit Programme 2018 report on 2017 data

Source of standard: National standard (RCPCH, RCOphth, BAPM and Bliss, Guideline for the Screening and Treatment of Retinopathy of Prematurity, 200829).

Outlier analysis: Yes

Inclusion criteria

Babies were included for analysis if they met the following criteria:

• Final neonatal discharge in the calendar year of analysis • Had care provided by an NNAP unit • Alive at the beginning of the national guideline screening window and

• The baby was born at less than 32 weeks gestational age and was admitted to a neonatal unit or

• The baby’s birth weight was less than 1501g.

Results

There were 9,681 babies born with a birth weight less than 1501g or with a gestational age at birth less than 32 weeks in a NNAP contributing neonatal unit. Of these babies, 17 were excluded because they did not have a recorded episode of care in a neonatal unit until after the closure of the ROP screening window. 22 babies were removed, as a responsible unit could not be assigned. A further 33 babies were excluded because they were transferred to non-neonatal units before, or during, the ROP screening window. Finally, 610 babies were excluded because they died before the closure of the screening window and had not been screened. This left 8,999 babies eligible for ROP screening from 179 neonatal units.

Including post-discharge screenings, 98.1% of eligible babies had at least one screening for ROP recorded, while 94.4% of babies were screened on-time in accordance with current NNAP criteria. Of the remaining babies, 3.1% were first screened after the closure of the screening window, and less than 1% were screened before the screening window opened. No screening data was available for 1.9% of eligible babies.

103

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.7.1: Timing of ROP screening, by neonatal unit level.

Screening time Any On time No NNU screen During After On time Early Late Screen Level NNU Babies (%) care discharge total (%) (%) (%) (%) 718 676 11 31 SCU 37 740 565 111 22 (3%) (97%) (91.4%) (1.5%) (4.2%) 3,645 3,504 24 117 LNU 88 3,716 2,983 521 71 (1.9%) (98.1%) (94.3%) (0.6%) (3.1%) 4,465 4,312 22 131 NICU 54 4,543 3,825 487 78 (1.7%) (98.3%) (94.9%) (0.5%) (2.9%) 8,828 8,492 57 279 171 Total 179 8,999 7,373 1,119 (98.1%) (94.4%) (0.6%) (3.1%) (1.9%)

Figure 5.7.1: Caterpillar plot of the rates of compliance for on-time ROP screening; neonatal units.

Rates of compliance with the standard for on-time ROP screening. The 95% confidence intervals are indicated by vertical bars. The units can be identified in NNAP Online.

104

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.7.2: Timing of ROP screening, by neonatal network.

Screening time No screen Network Babies Any screen (%) On time On time Early (%) Late (%) (%) During care After discharge total (%) East of England Neonatal 744 727 (97.7%) 615 76 691 (92.9%) 7 (0.9%) 29 (3.9%) 17 (2.3%) ODN Midlands South West 410 406 (99%) 335 50 385 (93.9%) 4 (1%) 17 (4.1%) 4 (1%) Newborn Neonatal ODN North Central & North East London Neonatal 683 657 (96.2%) 578 65 643 (94.1%) 0 (0%) 14 (2%) 26 (3.8%) ODN North West London 414 406 (98.1%) 306 80 386 (93.2%) 1 (0.2%) 19 (4.6%) 8 (1.9%) Neonatal ODN North West Neonatal 1019 1046 1040 (99.4%) 859 160 3 (0.3%) 18 (1.7%) 6 (0.6%) ODN (97.4%) Northern Neonatal ODN 342 326 (95.3%) 247 47 294 (86%) 2 (0.6%) 30 (8.8%) 16 (4.7%) Scotland 530 522 (98.5%) 443 53 496 (93.6%) 4 (0.8%) 22 (4.2%) 8 (1.5%) South East Coast 631 623 (98.7%) 521 84 605 (95.9%) 2 (0.3%) 16 (2.5%) 8 (1.3%) Neonatal ODN South London Neonatal 568 562 (98.9%) 445 81 526 (92.6%) 6 (1.1%) 30 (5.3%) 6 (1.1%) ODN South West Neonatal 551 546 (99.1%) 475 57 532 (96.6%) 1 (0.2%) 13 (2.4%) 5 (0.9%) ODN Staffordshire, Shropshire and Black Country 317 315 (99.4%) 279 32 311 (98.1%) 0 (0%) 4 (1.3%) 2 (0.6%) Neonatal ODN Thames Valley & Wessex 761 753 (98.9%) 634 95 729 (95.8%) 10 (1.3%) 14 (1.8%) 8 (1.1%) ODN Trent Perinatal & Central 734 698 (95.1%) 579 92 671 (91.4%) 9 (1.2%) 18 (2.5%) 36 (4.9%) Newborn Neonatal ODN Wales 389 375 (96.4%) 325 30 355 (91.3%) 4 (1%) 16 (4.1%) 14 (3.6%) Yorkshire & Humber 873 867 (99.3%) 728 117 845 (96.8%) 4 (0.5%) 18 (2.1%) 6 (0.7%) Neonatal ODN Isle of Man 6 5 (83.3%) 4 0 4 (66.7%) 0 (0%) 1 (16.7%) 1 (16.7%) 8492 Total 8999 8828 (98.1%) 7373 1119 57 (0.6%) 279 (3.1%) 171 (1.9%) (94.4%) 105

National Neonatal Audit Programme 2018 report on 2017 data

Figure 5.7.2: Caterpillar plot of the rates of compliance for on-time ROP screening; neonatal networks.

Rates of compliance with the standard for on-time ROP screening. The 95% confidence intervals are indicated by vertical bars. The networks are presented in the ascending order of the rates in 2017.

Table 5.7.3: Timing of ROP screening, by NNAP reporting year (2009–2017).

Screening time Any No NNAP NNU Babies screen On time screen year During After On time Early (%) Late (%) (%) (%) care discharge total (%) 2,098 1,859 1,379 2,577 2009 167 7,913 5,336 - - (26.5%) (23.5%) (17.4%) (32.6%) 4,455 297 1,127 2,399 2010 171 8,278 5,879 - - (53.8%) (3.6%) (13.6%) (29%) 5,310 917 1,427 2011 164 7,887 6,460 - - 233 (3%) (67.3%) (11.6%) (18.1%) 5,319 871 1,684 2012 173 7,996 6,312 4,842 477 122 (1.5%) (66.5%) (10.9%) (21.1%) 6,995 432 503 2013 175 8,000 7,497 6,25s8 737 70 (0.9%) (87.4%) (5.4%) (6.3%) 7,653 283 227 2014 173 8,224 7,997 6,723 930 61 (0.7%) (92.8%) (3.4%) (2.8%) 8,226 324 217 2015 179 8,821 8,604 7,138 1,088 54 (0.6%) (93.3%) (3.7%) (2.5%) 8,968 8,597 333 163 2016 181 9,131 7,456 1,141 38 (0.4%) (98.2%) (94.2%) (3.6%) (1.8%) 8,828 8,492 279 171 2017 179 8,999 7,373 1,119 57 (0.6%) (98.1%) (94.4%) (3.1%) (1.9%)

106

National Neonatal Audit Programme 2018 report on 2017 data

5.8. Encephalopathy

The key findings and recommendations for this audit measure are found on page 39.

NNAP audit measure

Does an admitted baby born at 35 weeks gestational age or above have an encephalopathy within the first three full calendar days after birth?

Change to the audit measure for the 2017 data year: None.

NNAP standard

Benchmarking.

Inclusion criteria

All babies born at 35 weeks gestational age or above within the year of analysis, regardless of neonatal admission, will be included as the denominator for this question. Details on this denominator will be obtained externally to the NNRD, which typically forms the denominator for NNAP audit measure (see data sources for more details).

107

National Neonatal Audit Programme 2018 report on 2017 data

Results

There were 2,060,531 babies born at greater than or equal to 35 weeks gestation between 01 January 2014 and 31 December 2016. Of these 3,372 were recorded as having an encephalopathy within three days of birth. Encephalopathy occurred in 1.64 babies per 1000 births (95% confidence intervals: 1.69 – 1.58).

Live birth data was not available for the year 2014 for one trust, and two trusts had no live birth data for any years.

Table 5.8.1: Encephalopathy rates per 1000 births.

Encephalopathy rate per 1000 Missing All live births Live births ≥ 35 weeks Encephalopathy births (95% CI) data (%) 657 2,181,353 2,060,531 3,372 1.64 (1.69-1.58) (0.03%)

108

National Neonatal Audit Programme 2018 report on 2017 data

5.9. Bloodstream infection

Sick and premature babies are prone to infection with germs including some that are normally harmless to healthy people. Infections can lengthen the stay in the neonatal unit and may worsen the long term developmental outlook for babies.30 Neonatal unit staff and parents can reduce the risk of infection by following good infection prevention and control practice.

The key findings and recommendations for this audit measure are found on page 40.

NNAP audit measure

What percentage of babies admitted to a neonatal unit have:

• one or more episodes of a pure growth of a pathogen from blood; • one or more episode of a pure growth of a pathogen from CSF; • either a pure growth of indeterminate significance or a mixed growth with three or more clinical signs at the time of blood sampling?

Change to the audit measure for the 2017 data year: For the 2017 data report, the list of organisms of which a growth is regarded as unequivocal evidence of infection has been reviewed, and can be found in Appendix F: “Pathogens” in the NNAP.

For the 2018 data year, cerebrospinal fluid (CSF) cultures will be removed from the audit measure, Data will be presented on bloodstream infection, without reference to the presence of symptoms and signs.

NNAP standard

Benchmarking.

109

National Neonatal Audit Programme 2018 report on 2017 data

Inclusion criteria

Babies were included for analysis if they met the following criteria:

• Final neonatal discharge in the calendar year of analysis • Had care provided by an NNAP unit.

Results

Overall, 70,066 blood cultures were reported from 104,183 babies in 179 neonatal units, giving an average of less than 1 culture per baby. Of these blood cultures, 81.4% have a result entered. Less than half of blood cultures have symptoms and signs entered into the audit.

A total of 75,935 blood and CSF cultures were recorded for these babies; pathogens results, including ‘no growth’ were entered for 81.1% of these known culture samples.

At lower gestations, 16,016 blood cultures were reported from 8,394 babies (birth gestational age of less than 32 weeks), while at higher gestations 54,045 blood cultures were reported from 95,727 babies. On aggregate blood culture results are recorded for 81.4% of entered blood cultures. However, the proportion of positive blood cultures whose results are recorded remains unknown, except in the 74 neonatal units who have confirmed they have entered 100% of all positive blood cultures.

Of all blood cultures entered in the audit, 49.7% have clinical symptoms and signs data entered.

Table 5.9.1: Completeness of blood and CSF cultures, by gestational age group.

Blood cultures CSF cultures Gestational Babies Number Results Clinical signs Number Results age group entered entered (%) entered (%) entered entered (%)

≤ 27 weeks 2,563 7,716 6,275 (81.3%) 3,501 (45.4%) 632 513 (81.2%) 28-31 5,831 8,300 6,835 (82.3%) 4,111 (49.5%) 434 332 (76.5%) weeks 32-36 29,745 19,439 15,927 (81.9%) 9,888 (50.9%) 767 570 (74.3%) weeks 28,006 ≥ 37 weeks 65,982 34,606 17,355 (50.2%) 4,036 3,138 (77.8%) (80.9%) Missing 62 5 4 (80%) 2 (40%) 0 0 (0%) 57,047 Total 104,183 70,066 34,857 (49.7%) 5,869 4,553 (77.6%) (81.4%)

110

National Neonatal Audit Programme 2018 report on 2017 data

These findings make clear that inter-unit comparisons based on NNAP reported rates of bloodstream infections and central line associated bloodstream infection (CLABSI) can only be made with any level of confidence for the units with known 100% entry of positive cultures, and that any measures of infection depending on entry completeness of symptoms and signs should be interpreted with caution.

• 985 babies had blood culture results recorded with a pure growth of a pathogen.

• 14 babies had one or more positive CSF culture result recorded with a pure growth of a pathogen.

• For blood cultures, 51 babies had a growth of indeterminate significance with three or more clinical predefined clinical signs, and 17 a mixed growth with three or more predefined clinical signs.

For the least mature babies, born at less than 28 weeks gestational age, confirmed infection rates appear high. 2563 babies had 521 growths of a pathogen – confirming the clinical importance of infection in this patient group. Under reporting remains a concern. This number of positive growths is much higher than in 2016 (2016 - 276) and therefore it remains possible that higher overall rates of infection might be reported in the event of complete capture of all positive organisms from all units.

Table 5.9.2: Positive blood cultures, by gestational age group.

Positive blood cultures One or more One or more One or more Gestational skin mixed Babies Admissions pure growths age group commensal growths and of a pathogen growths and ≥ ≥ 3 clinical (%) 3 clinical signs signs ≤ 27 weeks 2563 5536 521 (20.3%) 31 12

28-31 weeks 5831 8915 246 (4.2%) 13 5

32-36 weeks 29745 33612 112 (0.4%) 5 0 ≥ 37 weeks 65982 70480 106 (0.2%) 2 0 Missing 62 63 0 (0%) 0 0 Total 104183 118606 985 (0.9%) 51 17

111

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.9.3: Positive CSF cultures, by gestational age group.

Positive CSF cultures One or more pure Gestational age growths of a pathogen group Babies Admissions (%) ≤ 27 weeks 2563 5536 6 (0.2%) 28-31 weeks 5831 8915 2 (0.03%) 32-36 weeks 29745 33612 2 (0.006%) ≥ 37 weeks 65982 70480 4 (0.006%) Missing 62 63 0 (0%) Total 104183 118606 14 (0.01%)

112

National Neonatal Audit Programme 2018 report on 2017 data

5.10. Quality Improvement Surveillance Definition: Central line associated bloodstream infection

The key findings and recommendations for this audit measure are found on page 40.

NNAP audit measure

How many babies have a positive blood culture (any species) with a central line present, after the first 72 hours of life, per 1000 central line days?

Change to the audit measure for the 2017 data year: None.

NNAP standard

Benchmarking.

Inclusion criteria

Babies will be included for analysis if they meet the following criteria:

• Experienced their final neonatal discharge in the calendar year of analysis • Had care provided by an NNAP unit • All days where a central line (surgical venous line, umbilical venous catheter (UVC), umbilical arterial catheter (UAC), peripherally inserted central catheter (PICC)) was present will be included in the number of line days when calculating number of babies experiencing one or more bloodstream infections per 1000 line days.

113

National Neonatal Audit Programme 2018 report on 2017 data

Results

104,183 babies in 179 neonatal units received 1,190,039 days of care. In total 13.8% of all care days included a central line and 756 babies with 921 bloodstream infections (definition outlined above) were reported for these central line days; 4.61 babies with bloodstream infections per 1000 central line days (5.85 in babies less than 32 weeks gestational age, 2.15 in babies greater than or equal to 32 weeks gestational age).

In the 74 units with complete data entry, central line associated bloodstream infection occurred in 8.17 babies of less than 32 weeks gestational age per 1000 line days and in 2.84 babies of greater than or equal to 32 weeks gestational age per 1000 line days.

Table 5.10.1: Babies with central line associated bloodstream infections, by gestational age group.

Babies with Babies with central central line line associated Gestational age group Babies Line days associated bloodstream blood stream infection per 1000 infections central line days < 32 weeks 8,394 109,080 638 5.85 ≥ 32 weeks 95,727 54,832 118 2.15 Missing 62 2 0 0.00 Total 104,183 163,914 756 4.61

114

National Neonatal Audit Programme 2018 report on 2017 data

5.11. Central line associated bloodstream infection

NNAP audit measure

How many blood stream infectionsa are there on the neonatal unit per 1000 central lineb days? a: the growth of a recognised pathogen in pure culture, or in the case of a mixed growth, or growth of indeterminate significance, the added requirement for 3 or more of 10 predefined clinical signs b: central line = umbilical artery catheter(UAC), umbilical venous catheter (UVC), percutaneous long line or surgically inserted long line.

Change to the audit measure for the 2017 data year: The list of organisms regarded as indicative of infection without the need for confirmatory symptoms and signs has been revised with expert assistance and can be found in Appendix F: “Pathogens” in the NNAP.

For the 2018 data year, this measure will not be included.

NNAP standard

Benchmarking.

Inclusion criteria

Babies will be included for analysis if they meet the following criteria:

• Experienced their final neonatal discharge in the calendar year of analysis • Had care provided by an NNAP unit • All days where a central line (surgical venous line, umbilical venous catheter (UVC), umbilical artery catheter (UAC), peripherally inserted central catheter (PICC)) was present will be included in the number of line days.

115

National Neonatal Audit Programme 2018 report on 2017 data

Results

104,183 babies in 179 neonatal units received 1,190,039 days of care. In total 13.8% of all care days included a central line and 729 bloodstream infections (definition as outlined above) were reported for these central line days.

A minimum estimate for the number of infections per 1000 days of central line care is 4.45 (5.68 for babies less than 32 weeks gestation; 1.99 for babies of 32 weeks or more gestation). However, such an estimate is potentially undermined both by incomplete blood culture data in some units, but also by incomplete data on “symptoms and signs” (Table 5.9.1).

Table 5.11.1: Central line associated bloodstream infections, by gestational age group.

Central line Central line associated Gestational age associated Babies Line days bloodstream group blood stream infection per 1000 infections central line days < 32 weeks 8,394 109,080 620 5.68 ≥ 32 weeks 95,727 54,832 109 1.99 Missing 62 2 0 0 Total 104,183 163,914 729 4.45

116

National Neonatal Audit Programme 2018 report on 2017 data

5.12. Bronchopulmonary dysplasia (BPD)

Variations in rates of BPD might reflect different management or could reflect the way that neonatal units use oxygen in most mature babies.

The key findings and recommendations for this audit measure are found on page 43.

NNAP audit measure

Does an admitted baby born at less than 32 weeks develop bronchopulmonary dysplasia (BPD)?

Change to the audit measure for the 2017 data year: None.

For the 2018 data year, the NNAP is amending this measure in line with published evidence that oxygen or dependence on respiratory support at 36 weeks gestational age better predicts longer-term lung disease.31 This pragmatic definition of BPD, which does not take account of oxygen dependence in the first 28 days, is widely used in clinical trial and other academic work to describe this outcome.

NNAP standard

Benchmarking.

117

National Neonatal Audit Programme 2018 report on 2017 data

Inclusion criteria

Babies will be included for the BPD analysis if they meet the following criteria:

• Gestational age at birth is less than 32 weeks and

• The baby was still an inpatient in a neonatal unit at 36 weeks postmenstrual age or had been discharged alive from neonatal care at less than 36 weeks postmenstrual age.

Results

There were 25,418 babies born less than 32 weeks, discharged between 01 January 2015 and 31 December 2017 as reported by 190 neonatal units who were considered eligible for this audit measure. Of these babies, 901 were excluded as the complete data required for analysis of BPD was not available from units participating in the NNAP. In total 24,517 babies were eligible for inclusion in the analysis.

Over 3 years 52.3% babies were assessed as not having BPD, whilst 16.9% of babies were defined as having mild BPD and 30.9% were categorised as having significant BPD. BPD could not be determined for 0.2% of babies. 1,880 babies died before they reached 36 weeks corrected gestational age.

All babies were assigned to their recorded place of birth for this analysis. For the following tables responses are assigned “Other” if the mother was recorded as delivering the baby at home, in transit, in an unknown location or in a maternity unit not allied with a NNAP participating unit in the first neonatal unit admission.

118

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.12.1: BPD only, by neonatal unit level.

Babies alive at BPD Status Deaths before 36 weeks with Significant Missing data NNU level NNU Babies 36 weeks cga sufficient data Significant BPD No BPD Mild BPD BPD (%) (%) to attribute or death (%) (%) BPD outcome SCU 45 1,324 93 (7%) 1,228 803 173 252 (20.5%) 345 (26.1%) 3 (0.2%) 1,889 LNU 89 8,833 470 (5.3%) 8,344 5,042 1,413 2,359 (26.8%) 19 (0.2%) (22.6%) 4,830 NICU 56 14,360 1,317 (9.2%) 13,023 5,966 2,227 6,147 (42.9%) 20 (0.1%) (37.1%) 11,811 3,813 6,971 Total 190* 24,517 1,880 (7.7%) 22,595 8,851 (36.2%) 42 (0.2%) (52.3%) (16.9%) (30.9%) Other 31 420 68 (16.2%) 352 182 59 111 (31.5%) 179 (42.6%) 0 (0%) Isle of Man 1 8 0 (0%) 8 6 1 1 (12.5%) 1 (12.5%) 0 (0%)

*Number of neonatal units higher than number participating in 2017 due to three-year data.

119

National Neonatal Audit Programme 2018 report on 2017 data

Figure 5.12.1: Caterpillar plot of the rates of significant BPD or death; neonatal units.

Rates of significant BPD or death with the national rate of BPD or death. The 95% confidence intervals are indicated by vertical bars. The units can be identified in NNAP Online.

120

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.12.2: BPD only, by neonatal network.

Babies alive at 36 BPD status Deaths weeks with Significant before 36 Network Babies sufficient data to Mild Significant BPD or Missing data (%) weeks cga No BPD attribute BPD BPD BPD (%) death (%) (%) outcome East of England Neonatal ODN 1,923 109 (5.7%) 1,814 1,055 313 446 (24.6%) 555 (28.9%) 0 (0%) Midlands South West Newborn 1,147 124 (10.8%) 1,022 540 174 308 (30.1%) 432 (37.7%) 1 (0.1%) Neonatal ODN North Central & North East London 1,797 96 (5.3%) 1,690 848 310 532 (31.5%) 628 (35.2%) 11 (0.6%) Neonatal ODN North West London Neonatal ODN 1,124 80 (7.1%) 1,033 535 163 335 (32.4%) 415 (37.3%) 11 (1%) North West Neonatal ODN 3,173 303 (9.5%) 2,869 1,448 432 989 (34.5%) 1,292 (40.7%) 1 (0%) Northern Neonatal ODN 1,080 79 (7.3%) 996 479 137 380 (38.2%) 459 (42.7%) 5 (0.5%) Scotland 1,266 111 (8.8%) 1,155 570 203 382 (33.1%) 493 (38.9%) 0 (0%) South East Coast Neonatal ODN 1,732 117 (6.8%) 1,615 883 286 446 (27.6%) 563 (32.5%) 0 (0%) South London Neonatal ODN 1,593 111 (7%) 1,482 744 227 511 (34.5%) 622 (39.0%) 0 (0%) South West Neonatal ODN 1,514 101 (6.7%) 1,407 698 268 441 (31.3%) 542 (35.8%) 6 (0.4%) Staffordshire, Shropshire and Black 901 110 (12.2%) 791 396 159 236 (29.8%) 346 (38.4%) 0 (0%) Country Neonatal ODN Thames Valley & Wessex ODN 2,040 119 (5.8%) 1,919 1,053 308 558 (29.1%) 677 (33.2%) 2 (0.1%) Trent Perinatal & Central Newborn 1,895 164 (8.7%) 1,727 897 312 518 (30%) 682 (36.1%) 4 (0.2%) Neonatal ODN Wales 990 64 (6.5%) 925 467 179 279 (30.2%) 343 (34.7%) 1 (0.1%) Yorkshire & Humber Neonatal ODN 2,342 192 (8.2%) 2,150 1,198 342 610 (28.4%) 802 (34.2%) 0 (0%) 6971 Total 24,517 1880 (7.7%) 22,595 11,811 3813 8,851 (36.2%) 42 (0.2%) (30.9%) Other 420 68 (16.2%) 352 182 59 111 (31.5%) 179 (42.6%) 0 (0%) Isle of Man 8 0 (0%) 8 6 1 1 (12.5%) 1 (12.5%) 0 (0%)

121

National Neonatal Audit Programme 2018 report on 2017 data

Figure 5.12.2: Caterpillar plot of rates of BPD or death in very preterm babies cared for in neonatal networks in England, Wales and Scotland 2015-2017.

Rates of significant BPD or death with the national rate for BPD or death. The 95% confidence intervals are indicated by vertical bars. The networks are presented in the ascending order of the rates in 2017.

Figure 5.12.3: Caterpillar plot of ‘treatment effect’ on rates of BPD or death in very preterm babies cared for in neonatal networks in England, Wales and Scotland 2015-2017.

“Treatment effect” is the difference between the rate of BPD or death in babies cared for in a neonatal network compared to the observed rate for a matched group of babies with very similar case mix, cared for in all neonatal units. A positive treatment effect indicates that the rate of BPD or death is higher in the network of interest than for a comparable

122

National Neonatal Audit Programme 2018 report on 2017 data group of babies cared for in all neonatal units. Where the 95% confidence interval for this effect excludes 0, the treatment effect is unlikely to be a chance finding.

Explaining ‘treatment effect’

As an example, consider the combined rate of BPD or death in the Northern region. It is the highest of the networks’ rates, and the 95% confidence intervals indicate that this is not a chance finding. However, within the network treatment effect analysis, babies cared for in the Northern region had comparable rates of BPD or death compared to those cared for in all participating units – and therefore only a small treatment effect, whose 95% confidence intervals cross zero. Therefore it is likely that explanations other than how babies are cared for in the Northern region explain the high reported rates of BPD or death.

By contrast the upper panel shows that the combined rate of BPD or death in the South East Coast Neonatal Network is lower than the national rate. When the rate of BPD or death for a set of babies matched to those cared for in the South East Coast Neonatal Network is compared, a negative treatment effect is observed, with 95% confidence intervals excluding zero. This suggests that treatment in South East Coast Neonatal Network is associated with 3.9% lower rates of BPD or death.

Table 5.12.3: Rates of BPD only, by NNAP reporting period (2013–2017).

With Significant BPD Missing NNAP Year NNU Babies outcome (%) data (%)

2013-2015 182 21,805 21,673 6,508 (30%) 132 (0.6%)

2014-2016 183 22,049 21,978 6,792 (30.9%) 71 (0.3%)

2015-2017 190 24,517 22,595 6,971 (30.9%) 42 (0.2%)

123

National Neonatal Audit Programme 2018 report on 2017 data

5.13. Necrotising enterocolitis

The key findings and recommendations for this audit measure are found on page 45.

NNAP audit measure

Does an admitted baby born at less than 32 weeks gestational age meet the NNAP surveillance definition for necrotising enterocolitis (NEC) on one or more occasion?

For this outcome, babies are assigned to the unit of presence at the age of 48 hours as a proxy measure of the unit that was intended to provide ongoing care for them.

Change to the audit measure for the 2017 data year: New measure for the 2017 data year.

NNAP standard

Benchmarking.

Inclusion criteria

Babies were included for analysis if they meet the following criteria:

• Final neonatal discharge in the calendar year of analysis • Care provided by an NNAP unit • Born at less than 32 weeks gestational age and survived to at least 48 hours after birth • Admitted for one or more episodes of care in a neonatal unit.

124

National Neonatal Audit Programme 2018 report on 2017 data

Results

8,228 babies were born at less than 32 weeks and survived to 48 hours after birth. Of these, 5.6% had a confirmed case of NEC on one or more occasion.

For 600 babies it was not possible to determine whether the baby had NEC at any point in their neonatal care.

Table 5.13.1: NEC, by neonatal unit level.

NEC Status Missing data

Died prior to Death NNU With Alive at NNU Babies discharge No before level outcome NEC discharge home, but no NEC discharge (%) NEC (%) SCU 33 186 174 0 171 3 (1.7%) 11 (5.9%) 1 (0.5%) LNU 85 2,542 2,304 16 2,220 68 (3%) 234 (9.2%) 4 (0.2%) 354 NICU 54 5,469 5,122 267 4,501 242 (4.4%) 105 (1.9%) (6.9%) 3 Other - 31 28 0 25 3 (9.7%) 0 (0%) (10.7%) 428 Total 172 8,228 7,628 283 6,917 490 (6%) 110 (1.3%) (5.6%)

125

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.13.2: NEC, by neonatal network.

NEC status Missing data With Died prior to Network Babies No Alive at Death before outcome discharge home, but NEC NEC discharge (%) discharge (%) no NEC East of England Neonatal ODN 633 607 18 547 42 (6.9%) 26 (4.1%) 0 (0%) Midlands South West Newborn 356 324 13 288 23 (7.1%) 23 (6.5%) 9 (2.5%) Neonatal ODN North Central & North East London 574 545 16 500 29 (5.3%) 28 (4.9%) 1 (0.2%) Neonatal ODN North West London Neonatal ODN 363 350 13 321 16 (4.6%) 8 (2.2%) 5 (1.4%) North West Neonatal ODN 1,039 952 44 864 44 (4.6%) 53 (5.1%) 34 (3.3%) Northern Neonatal ODN 307 250 5 233 12 (4.8%) 49 (16%) 8 (2.6%) Scotland 495 457 20 416 21 (4.6%) 31 (6.4%) 7 (1.4%) South East Coast Neonatal ODN 587 549 19 493 37 (6.7%) 34 (5.8%) 4 (0.7%) South London Neonatal ODN 514 468 17 402 49 (10.5%) 40 (7.8%) 6 (1.2%) South West Neonatal ODN 562 496 22 458 16 (3.2%) 64 (11.4%) 2 (0.4%) Staffordshire, Shropshire and Black 292 279 15 247 17 (6.1%) 10 (3.4%) 3 (1%) Country Neonatal ODN Thames Valley & Wessex ODN 711 675 23 622 30 (4.4%) 27 (3.8%) 9 (1.3%) Trent Perinatal & Central Newborn 598 544 18 485 41 (7.5%) 46 (7.7%) 8 (1.3%) Neonatal ODN Wales 333 323 7 299 17 (5%) 6 (1.8%) 4 (1.2%) Yorkshire & Humber Neonatal ODN 830 778 33 714 31 (4%) 42 (5.1%) 10 (1.2%) Isle of Man 3 3 0 3 0 (0%) 0 (0%) 0 (0%) Other 31 28 0 24 3 (10.7%) 3 (9.7%) 0 (0%) Total 8,228 7,628 283 6,917 428 (5.6%) 490 (6%) 110 (1.3%)

126

National Neonatal Audit Programme 2018 report on 2017 data

5.14. Minimising separation of mother and term baby

Some babies admitted to neonatal units may be separated from their mothers for longer than necessary. It may be possible to care for some babies in transitional care, a setting which takes an interdisciplinary approach of both midwives and neonatal staff to deliver high-quality care to both mothers and babies and avoid their separation. 32 This measure seeks to describe the number of babies admitted to neonatal units for low dependency care and to compare the number of days that babies were separated from their mothers.

The key findings and recommendations for this audit measure are found on page 47.

NNAP audit measure

For a baby born at gestational age greater than or equal to 37 weeks, who did not have any surgery or a transfer during any admission, how many special carea or normal careb days were provided when oxygen was not administered? a= Healthcare Resource Group (HRG) 3, or b= HRG 5, as defined by the NHS England neonatal critical care service specification.24

Change to the audit measure for the 2017 data year: New measure for the 2017 data year.

NNAP standard

Benchmarking.

127

National Neonatal Audit Programme 2018 report on 2017 data

Inclusion criteria

Babies were included for analysis if they met the following criteria:

• Gestational age at birth greater than or equal to 37 weeks • Received all their care in one NNAP unit • Admitted for at least 12 hours • Did not have major surgery • Had care given on a neonatal unit at any point during the admission

Normal or special care days for these eligible babies were counted where they met the following criteria:

• Nursed on a neonatal unit on that day, or days • No oxygen or other form of non-invasive respiratory support was provided on that day or days • HRG level was 3 or 5

Results

Almost all (28,524 of 31,725, 89.9%) of admitted term babies, who did not have surgery and were not transferred, had some special or normal care days on which oxygen was not administered. 100,771 special care and normal care days (67,069 special care; 33,702 normal care) were provided to these 28,524 babies. On average 3.2 special care or normal care days were given for each neonatal unit admitted baby in this gestation category.

The NNAP notes that some neonatal units provide in house accommodation for mother and baby, which, in the absence of midwifery input, is not currently recorded as neonatal transitional care; this may result in some local over estimation of term baby separation.

128

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.14.1: Babies spending one or more days in special or normal care, by neonatal unit level.

Number of eligible care Babies who days received one Number Total or more of NNU special NNU Babies eligible days separation level Special Normal and in special or days per care care normal normal care baby care (%) days SCU 37 3,667 3,374 (92%) 7,648 4,531 12,179 3.3 LNU 88 13,958 12,791 (91.6%) 29,074 15,524 44,598 3.2 NICU 54 14,100 12,359 (87.7%) 30,347 13,647 43,994 3.1 Total 179 31,725 28,524 (89.9%) 67,069 33,702 100,771 3.2

129

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.14.2: Babies spending one or more days in special or normal care, by neonatal network.

Number of eligible care days Babies who received one Number of Network Babies or more eligible days in Total special and separation special or normal care (%) Special care Normal care normal care days per baby days East of England Neonatal ODN 4,110 3,716 (90.4%) 8,654 3,855 12,509 3 Midlands South West Newborn 1,841 1,560 (84.7%) 3,181 2,040 5,221 2.8 Neonatal ODN North Central & North East 2,570 2,379 (92.6%) 6,464 2,795 9,259 3.6 London Neonatal ODN North West London Neonatal 994 883 (88.8%) 2,023 1,116 3,139 3.2 ODN North West Neonatal ODN 3,367 3,195 (94.9%) 8,513 4,167 12,680 3.8 Northern Neonatal ODN 1,053 943 (89.6%) 2,190 1,253 3,443 3.3 Scotland 2,216 2,075 (93.6%) 4,758 3,168 7,926 3.6 South East Coast Neonatal ODN 1,899 1,664 (87.6%) 3,523 1,776 5,299 2.8 South London Neonatal ODN 2,006 1,844 (91.9%) 5,115 1,635 6,750 3.4 South West Neonatal ODN 2,530 2,086 (82.5%) 4,287 2,522 6,809 2.7 Staffordshire, Shropshire and 1,026 874 (85.2%) 1,954 1,083 3,037 3 Black Country Neonatal ODN Thames Valley & Wessex ODN 2,301 2,088 (90.7%) 4,880 2,178 7,058 3.1 Trent Perinatal & Central 2,223 2,057 (92.5%) 4,259 1,825 6,084 2.7 Newborn Neonatal ODN Wales 1,196 1,094 (91.5%) 2,457 1,442 3,899 3.3 Yorkshire & Humber Neonatal 2,330 2,005 (86.1%) 4,656 2,751 7,407 3.2 ODN Isle of Man 63 61 (96.8%) 155 96 251 4 Total 31,725 28,524 (89.9%) 67,069 33,702 100,771 3.2

130

National Neonatal Audit Programme 2018 report on 2017 data

5.15. Minimising separation of mother and late preterm baby (34-36 weeks)

The key findings and recommendations for this audit measure are found on page 47.

NNAP audit measure

For a baby born at 34-36 weeks gestational age, who did not have any surgery or a transfer during any admission, how many special care a or normal careb days were provided when oxygen was not administered? a= HRG 3 or b= HRG 5, as defined by the NHS England neonatal critical care service specification.24

Change to the audit measure for the 2017 data year: New measure for the 2017 data year.

NNAP standard

Benchmarking.

Inclusion criteria

Babies will be included for analysis if they meet the following criteria:

• Born between 34 and 36 weeks gestational age • Received all their care in one NNAP unit • Admitted for at least 12 hours • Did not have major surgery

Normal or special care days for these eligible babies will be counted where they meet the following criteria:

• Nursed on a neonatal unit on a day, or days • No oxygen or other form of non-invasive respiratory support was provided on that day or days

131

National Neonatal Audit Programme 2018 report on 2017 data

• HRG level is 3 or 5

Results

Almost all (14,917 of 15,649; 95.3%) of admitted babies born at 34-36 weeks gestation received one or more days of special or normal care on which oxygen was not administered. 105,723 special and normal care days (81,604 special care; 24,119 normal care) were provided, meaning that on average 6.8 special care or normal care days were given for each neonatal unit admitted baby in this gestation category.

Table 5.15.1: Babies spending one or more days in special or normal care, by neonatal unit level.

Number of eligible care Babies who days received one Number Total or more of NNU special NNU Babies eligible days separation level Special Normal and in special or days per care care normal normal care baby care (%) days SCU 37 2,021 1,975 (97.7%) 11,598 3,727 15,325 7.6 LNU 88 7,496 7,235 (96.5%) 39,298 12,564 51,862 6.9 NICU 54 6,132 5,707 (93.1%) 30,708 7,828 38,536 6.3 Total 179 15,649 14,917 (95.3%) 81,604 24,119 105,723 6.8

132

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.15.2: Babies spending one or more days in special or normal care, by neonatal network.

Number Babies who received Number of eligible care days of one or more eligible Network Babies Total special separation days in special or Special Normal and normal days per normal care (%) care care care days baby East of England Neonatal ODN 1,854 1,778 (95.9%) 8,578 2,258 10,836 5.8 Midlands South West Newborn 689 622 (90.3%) 2,968 935 3,903 5.7 Neonatal ODN North Central & North East London 1,005 978 (97.3%) 5,965 1,747 7,712 7.7 Neonatal ODN North West London Neonatal ODN 476 449 (94.3%) 2,529 770 3,299 6.9 North West Neonatal ODN 1,905 1,867 (98%) 11,545 3,758 15,303 8 Northern Neonatal ODN 628 609 (97%) 3,939 1,089 5,028 8 Scotland 1,211 1,173 (96.9%) 7,316 1,819 9,135 7.5 South East Coast Neonatal ODN 1,134 1,069 (94.3%) 5,529 1,428 6,957 6.1 South London Neonatal ODN 819 796 (97.2%) 4,503 1,021 5,524 6.7 South West Neonatal ODN 1,025 951 (92.8%) 5,390 1,805 7,195 7 Staffordshire, Shropshire and Black 528 481 (91.1%) 2,423 785 3,208 6.1 Country Neonatal ODN Thames Valley & Wessex ODN 1,115 1,078 (96.7%) 5,912 1,631 7,543 6.8 Trent Perinatal & Central Newborn 1,193 1,128 (94.6%) 4,688 1,449 6,137 5.1 Neonatal ODN Wales 601 579 (96.3%) 2,755 986 3,741 6.2 Yorkshire & Humber Neonatal ODN 1,437 1,330 (92.6%) 7,321 2,596 9,917 6.9 Isle of Man 29 29 (100%) 243 42 285 9.8 Total 15,649 14,917 (95.3%) 81,604 24,119 105,723 6.8

133

National Neonatal Audit Programme 2018 report on 2017 data

5.16. Breastmilk feeding at discharge home

The key findings and recommendations for this audit measure are found on page 50.

NNAP audit measure

Does a baby born at less than 33 weeks gestational age receive any of their own mother’s milk at discharge to home from a neonatal unit?28

Change to the audit measure for the 2017 data year: None.

NNAP standard

Benchmarking.

Inclusion criteria

Babies were included in the analysis if they met the following criteria:

• Born at less than 33 weeks gestational age • Received all their neonatal care in one neonatal unit, and were discharged home at the end of their neonatal care • Experienced their final neonatal discharge in the calendar year of analysis • Had care provided by an NNAP unit

134

National Neonatal Audit Programme 2018 report on 2017 data

Results

Of the 11,282 babies born at less than 33 weeks and admitted to an NNAP neonatal unit, there were 6,418 babies who met the criteria for inclusion in this question.

Daily data summaries for the last or penultimate day of care indicated that 60.5% of eligible babies were receiving mother’s milk, exclusively or with another form of feeding, at the time of their discharge from neonatal care. Of the remaining babies, 39.5% were recorded as receiving other types of feeding at discharge and 0.4% had no feeding data available from the last or penultimate day of care. Other types of enteral feeds are; ‘Formula’, ’Donor expressed breast milk’ and ‘Nil by mouth’.

This question is restricted to non-transferred babies, so that unit-level analysis can attribute this outcome solely to unit processes. However, in doing so 4,024 otherwise eligible babies are excluded from the analysis, which remains a limitation to the utility of this metric.

Table 5.16.1: Breastmilk feeding at discharge, by neonatal unit clinical level.

Enteral feeds at the time of NNU With discharge Missing NNU Babies level outcome Any breast No breast milk Data (%) milk (%) (%) SCU 34 341 336 230 (68.5%) 106 (31.5%) 5 (1.5%) LNU 88 2,851 2,841 1,761 (62%) 1,080 (38%) 10 (0.4%) NICU 54 3,226 3,217 1,875 (58.3%) 1,342 (41.7%) 9 (0.3%) 3,866 Total 176 6,418 6,394 2,528 (39.5%) 24 (0.4%) (60.5%)

Figure 5.16.1: Caterpillar plot of the rates of breastmilk feeding at discharge; neonatal units.

135

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.16.2: Breastmilk feeding at discharge, by neonatal network.

Enteral feeds at the time of With discharge Missing Network Babies outcome Any breast No breast Data (%) milk (%) milk (%) East of England 560 556 379 (68.2%) 177 (31.8%) 4 (0.7%) Neonatal ODN Midlands South West Newborn Neonatal 279 279 175 (62.7%) 104 (37.3%) 0 (0%) ODN North Central & North East London Neonatal 428 428 319 (74.5%) 109 (25.5%) 0 (0%) ODN North West London 273 271 238 (87.8%) 33 (12.2%) 2 (0.7%) Neonatal ODN North West Neonatal 747 743 368 (49.5%) 375 (50.5%) 4 (0.5%) ODN Northern Neonatal 213 212 109 (51.4%) 103 (48.6%) 1 (0.5%) ODN Scotland 457 456 239 (52.4%) 217 (47.6%) 1 (0.2%)

South East Coast 433 433 271 (62.6%) 162 (37.4%) 0 (0%) Neonatal ODN South London 344 339 264 (77.9%) 75 (22.1%) 5 (1.5%) Neonatal ODN South West Neonatal 429 427 264 (61.8%) 163 (38.2%) 2 (0.5%) ODN Staffordshire, Shropshire and Black 257 257 125 (48.6%) 132 (51.4%) 0 (0%) Country Neonatal ODN Thames Valley & 564 563 367 (65.2%) 196 (34.8%) 1 (0.2%) Wessex ODN Trent Perinatal & Central Newborn 507 504 270 (53.6%) 234 (46.4%) 3 (0.6%) Neonatal ODN Wales 250 250 131 (52.4%) 119 (47.6%) 0 (0%)

Yorkshire & Humber 672 671 345 (51.4%) 326 (48.6%) 1 (0.1%) Neonatal ODN Isle of Man 5 5 2 (40%) 3 (60%) 0 (0%) 2,528 24 Total 6,418 6,394 3,866 (60.5%) (39.5%) (0.4%)

136

National Neonatal Audit Programme 2018 report on 2017 data

Figure 5.16.2: Caterpillar plot of the rates of breastmilk feeding at discharge; neonatal networks.

Table 5.16.3: Breastmilk feeding at discharge, by NNAP reporting year (2013–2017).

Enteral feeds at the time of With discharge Missing NNAP Year NNU Babies outcome Any breast milk No breast milk data (%) (%) (%) 2013 170 5,920 5,902 3,509 (59.5%) 2,393 (40.5%) 18 (0.3%) 2014 169 5,942 5,866 3,570 (60.9%) 2,296 (39.1%) 76 (1.3%) 2015 175 6,323 6,268 3,693 (58.9%) 2,575 (41.1%) 55 (0.9%) 2016 176 6,574 6,473 3,866 (59.7%) 2,607 (40.3%) 101 (1.5%)

2017 176 6,418 6,394 3,866 (60.5%) 2,528 (39.5%) 24 (0.4%)

137

National Neonatal Audit Programme 2018 report on 2017 data

5.17. Follow-up at two years of age

The key findings and recommendations for this audit measure are found on page 52.

NNAP audit measure

Does a baby born at less than 30 weeks gestational age receive medical follow-up at two years corrected age (18-30 months gestationally corrected age)?

Does a baby have complete results of a structured assessment recorded24,33?

Babies are assigned to their final neonatal unit of discharge for this measure, on the basis that most babies are discharged from the centre which will be responsible for arranging follow up.

Change to the audit measure for the 2017 data year: None

NNAP standard

100% of babies with two-year follow-up data entered.

Source of standard: NNAP Project Board

Outlier analysis: Yes

138

National Neonatal Audit Programme 2018 report on 2017 data

Inclusion criteria

• Babies born at less than 30 weeks who are not recorded as deceased within their episodic data (including final neonatal outcome). • Babies whose parents or carers have not opted them out of secondary use of their data. • The eligible cohort runs for births from July 2014 to June 2015.

Results

There were 4,043 babies less than 30 weeks gestation born between July 2014 and June 2015 who survived and were discharged from a neonatal unit to home, to a ward or to foster care. Of these babies 62.6% had some/all health data entered. The remaining 1,512 (37.4%) babies had no two-year follow-up health data entered at all.

48.7% of babies who had two-year follow-up health data entered and did not die post discharge, had data entered in the standardised assessment sections (Bayley III, Griffiths or Schedule of Growing) of the two-year follow-up forms on BadgerNet.

Table 5.17.1: Two year follow up assessment for babies born between July 2014 and June 2015.

Year Babies Some health data entered (%) No health data entered (%) 2017 4,043 2,531 (62.6%) 1,512 (37.4%)

Figure 5.17.1: Caterpillar plot of the rates of two year follow up assessment; neonatal units.

139

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.17.2: Two year follow up assessment for babies born between July 2014 and June 2015, by neonatal network.

Some health No health data Network Babies data entered entered (%) (%) East of England Neonatal ODN 366 213 (58.2%) 153 (41.8%) Midlands South West Newborn 195 142 (72.8%) 53 (27.2%) Neonatal ODN North Central & North East 315 151 (47.9%) 164 (52.1%) London Neonatal ODN North West London Neonatal 155 86 (55.5%) 69 (44.5%) ODN North West Neonatal ODN 528 327 (61.9%) 201 (38.1%) Northern Neonatal ODN 169 99 (58.6%) 70 (41.4%) Scotland 188 128 (68.1%) 60 (31.9%) South East Coast Neonatal ODN 297 170 (57.2%) 127 (42.8%) South London Neonatal ODN 267 144 (53.9%) 123 (46.1%) South West Neonatal ODN 270 192 (71.1%) 78 (28.9%) Staffordshire, Shropshire and 119 82 (68.9%) 37 (31.1%) Black Country Neonatal ODN Thames Valley & Wessex ODN 319 233 (73%) 86 (27%) Trent Perinatal & Central 316 192 (60.8%) 124 (39.2%) Newborn Neonatal ODN Wales 153 104 (68%) 49 (32%) Yorkshire & Humber Neonatal 386 268 (69.4%) 118 (30.6%) ODN Total 4,043 2,531 (62.6%) 1,512 (37.4%)

140

National Neonatal Audit Programme 2018 report on 2017 data

Figure 5.17.2: Caterpillar plot of the rates of two year follow up assessment; neonatal networks.

Table 5.17.3: Two year follow up rates of compliance, by NNAP reporting year (2012-2017).

Year Babies Some health data entered (%) No health data entered (%)

2012 2,967 1,242 (41.9%) 1,725 (58.1%) 2013 3,488 1,561 (44.8%) 1,927 (55.2%) 2014 3,656 1.993 (54.5%) 1,663 (44.5%) 2015 3,744 2,265 (60.5%) 1,479 (39.5%) 2016 4,023 2,450 (60.9%) 1,573 (39.1%) 2017 4,043 2,531 (62.6%) 1,512 (37.4%)

141

National Neonatal Audit Programme 2018 report on 2017 data

Table 5.17.4: Assessments used for two-year follow up and network

Assessment used Babies with health data Bayley III, Griffiths or Network entered who did not Schedule of growing Schedule Other Unknown die post discharge assessment used (%) Bayley III Griffiths of growing

East of England Neonatal ODN 207 161 148 10 3 2 44

Midlands South West Newborn 141 65 4 0 61 0 76 Neonatal ODN North Central & North East London 148 87 82 4 1 3 58 Neonatal ODN

North West London Neonatal ODN 85 43 40 3 0 1 41

North West Neonatal ODN 319 27 18 0 9 9 283

Northern Neonatal ODN 98 61 58 0 3 0 37

Scotland 127 60 48 0 12 3 64

South East Coast Neonatal ODN 169 100 73 1 26 2 67

South London Neonatal ODN 143 98 63 18 17 0 45

South West Neonatal ODN 191 159 136 19 4 1 31

Staffordshire, Shropshire and Black 79 56 55 0 1 2 21 Country Neonatal ODN

Thames Valley & Wessex ODN 230 79 67 0 12 11 140

Trent Perinatal & Central Newborn 191 62 21 0 41 6 123 Neonatal ODN Wales 103 81 66 11 4 2 20

Yorkshire & Humber Neonatal ODN 259 73 1 0 72 4 182

Total 2,490 1,212 (48.7%) 880 66 266 46 1,232

142

National Neonatal Audit Programme 2018 report on 2017 data

Appendix A: Data completeness and unit level of participating units

For each NNAP audit measure, there is an associated rate of missing entries. The rate of missing entries is described as M/T, where M is the number of missing entries and T is the number of all cases. To summarise data completeness across all NNAP measures, an average was taken of the rates of missing entries associated with a given unit. This methodology was applied to the 2017 data shown in this report (see Table A).

ROP was not included in the data completeness summary measure as it is not possible to differentiate between a negative (“never screened”) outcome and missing data. Infection data is known to be incompletely entered on to the BadgerNet system and is omitted from the data completeness summary measure. In addition, it is not possible to calculate a rate of missing entries for the number of care days given for the minimising avoidable mother baby separation measures so these measures are also omitted.

For the NEC, BPD and Encephalopathy measures, the rate of missing entries is derived from the completeness of the daily or episodic data items used to categorise the results for these measures for the 2017 data year. This ensures units’ completeness is measures using the information they have responsibility for inputting into BadgerNet.

To calculate the number of eligible babies for one or more NNAP measures, a sum of all the eligible babies reported for each unit for the 2017 data year is taken.

Babies that are included in the Encephalopathy measure are not included in this count as it covers 2014-2016 data only. Similarly, only eligible babies for the 2017 data year for the BPD measure are included in this count.

For full details of unit level missing data, please see NNAP Online.

143

National Neonatal Audit Programme 2018 report on 2017 data

Table A: NNAP participating neonatal units1 (2017 data), average missing data and eligible babies.

Note that some units that are included in the Encephalopathy measure are not included in this table as they did not contribute 2017 data to the audit due to closure.

Number of Average Unit babies eligible Neonatal unit missing data level for one or more (%) NNAP measures ENGLAND East of England Neonatal ODN Bedford Hospital SCU 5.0% 312 Hinchingbrooke Hospital SCU 5.2% 335 James Paget Hospital SCU 9.3% 416 West Suffolk Hospital SCU 4.2% 361 Basildon Hospital LNU 10.0% 451 Broomfield Hospital LNU 7.3% 397 Colchester General Hospital LNU 9.2% 477 Ipswich Hospital LNU 15.3% 711 Lister Hospital LNU 2.7% 952 Peterborough City Hospital LNU 6.0% 1265 Princess Alexandra Hospital LNU 8.5% 485 Queen Elizabeth Hospital, King's Lynn LNU 4.2% 535 Southend Hospital LNU 7.8% 361 Watford General Hospital LNU 2.1% 1,426 Luton & Dunstable Hospital NICU 3.3% 995 Norfolk & Norwich University Hospital NICU 5.7% 1,297 Rosie Maternity Hospital, Addenbrookes NICU 4.9% 911 Midlands South West Newborn Neonatal ODN Good Hope Hospital SCU 10.3% 621 Hereford County Hospital SCU 3.7% 266 City Hospital, Birmingham LNU 8.8% 1,217 Worcestershire Royal Hospital LNU 5.6% 1,172 Birmingham Heartlands Hospital NICU 12.0% 1,335 Birmingham Women's Hospital NICU 9.1% 2,011 North Central & North East London Neonatal ODN The Royal Free Hospital SCU 3.3% 396 Barnet Hospital LNU 4.7% 1,608 Newham General Hospital LNU 15.1% 648 North Middlesex University Hospital LNU 8.0% 568 Queen's Hospital, Romford LNU 4.3% 937 Whipps Cross University Hospital LNU 22.9% 494 Whittington Hospital LNU 8.8% 2,188

1 Three eligible units did not participate in 2017; Dr Gray’s Hospital, Elgin, Simpsons Centre for Reproductive Health, Royal Infirmary of Edinburgh, and St John’s Hospital, Livingston.

144

National Neonatal Audit Programme 2018 report on 2017 data

Homerton Hospital NICU 4.8% 1,466 The Royal London Hospital NICU 21.8% 811 University College Hospital NICU 4.0% 971 North West London Neonatal ODN West Middlesex University Hospital SCU 10.7% 532 Hillingdon Hospital LNU 1.5% 774 Northwick Park Hospital LNU 11.0% 523 St Mary's Hospital, London LNU 16.5% 433 Chelsea & Westminster Hospital NICU 16.1% 806 Queen Charlotte's Hospital NICU 11.1% 548 North West Neonatal ODN Furness General Hospital SCU 2.1% 85 Countess of Chester Hospital LNU 5.6% 391 Leighton Hospital LNU 3.9% 294 Macclesfield District General Hospital LNU 2.5% 153 North Manchester General Hospital LNU 2.9% 541 Ormskirk District General Hospital LNU 10.0% 334 Royal Albert Edward Infirmary LNU 3.6% 334 Royal Lancaster Infirmary LNU 2.1% 234 Stepping Hill Hospital LNU 4.3% 361 Tameside General Hospital LNU 5.4% 284 Victoria Hospital, Blackpool LNU 3.6% 445 Warrington Hospital LNU 9.3% 506 Whiston Hospital LNU 9.5% 365 Wythenshawe Hospital LNU 1.3% 396 Arrowe Park Hospital NICU 8.5% 433 Lancashire Women & Newborn Centre NICU 7.1% 690 Liverpool Women's Hospital NICU 18.5% 3,238 Royal Bolton Hospital NICU 9.4% 676 Royal Oldham Hospital NICU 7.9% 746 Royal Preston Hospital NICU 7.6% 520 St Mary's Hospital, Manchester NICU 6.9% 1,262 Northern Neonatal ODN Cumberland Infirmary SCU 25.1% 199 Darlington Memorial Hospital SCU 13.1% 151 Northumbria Specialist Emergency Care SCU 434 6.7% Hospital Queen Elizabeth Hospital, Gateshead SCU 12.7% 223 South Tyneside District Hospital SCU 18.6% 100 University Hospital of North Durham SCU 12.8% 227 West Cumberland Hospital SCU 10.5% 135 James Cook University Hospital NICU 5.8% 475 Royal Victoria Infirmary NICU 12.5% 713 Sunderland Royal Hospital NICU 7.5% 300 University Hospital of North Tees NICU 6.7% 413

145

National Neonatal Audit Programme 2018 report on 2017 data

South East Coast Neonatal ODN Conquest Hospital SCU 2.8% 457 Darent Valley Hospital SCU 10.6% 809 Princess Royal Hospital SCU 8.5% 235 Queen Elizabeth The Queen Mother SCU 367 3.8% Hospital Royal Surrey County Hospital SCU 3.6% 546 Worthing Hospital SCU 1.9% 543 East Surrey Hospital LNU 7.3% 807 Frimley Park Hospital LNU 5.4% 906 Tunbridge Wells Hospital LNU 11.9% 698 Medway Maritime Hospital NICU 7.1% 1,148 Royal Sussex County Hospital NICU 9.1% 956 St Peter's Hospital NICU 3.2% 578 William Harvey Hospital NICU 2.0% 584 South London Neonatal ODN Epsom General Hospital SCU 5.0% 192 Princess Royal University Hospital SCU 22.5% 472 Croydon University Hospital LNU 10.9% 433 Kingston Hospital LNU 7.0% 425 Queen Elizabeth Hospital, Woolwich LNU 15.2% 425 St Helier Hospital LNU 3.6% 686 University Hospital Lewisham LNU 12.9% 471 Guy's & St Thomas' Hospital NICU 12.3% 1,079 King's College Hospital NICU 6.6% 769 St George's Hospital NICU 5.9% 2,731 South West Neonatal ODN North Devon District Hospital SCU 9.2% 253 Torbay Hospital SCU 4.0% 347 Yeovil District Hospital SCU 4.0% 228 Gloucestershire Royal Hospital LNU 6.6% 628 Great Western Hospital LNU 13.4% 784 Royal Cornwall Hospital LNU 2.5% 705 Royal Devon & Exeter Hospital LNU 2.5% 633 Royal United Hospital LNU 6.1% 691 Taunton & Somerset Hospital LNU 3.0% 597 Derriford Hospital NICU 6.8% 1,257 Southmead Hospital NICU 7.4% 3,041 St Michael's Hospital NICU 11.6% 3,055 Staffordshire, Shropshire and Black Country Neonatal ODN Manor Hospital LNU 4.9% 820 Princess Royal Hospital, Telford LNU 0.3% 1,207 Russells Hall Hospital LNU 6.1% 735 New Cross Hospital NICU 4.9% 1,342 Royal Stoke University Hospital NICU 3.7% 1,246

146

National Neonatal Audit Programme 2018 report on 2017 data

Thames Valley & Wessex ODN Dorset County Hospital SCU 2.9% 249 Basingstoke & North Hampshire Hospital LNU 3.0% 277 Milton Keynes Foundation Trust Hospital LNU 4.6% 395 Poole Hospital NHS Foundation Trust LNU 2.2% 445 Royal Berkshire Hospital LNU 3.4% 522 Royal Hampshire County Hospital LNU 3.4% 281 Salisbury District Hospital LNU 3.0% 287 St Mary's Hospital, IOW LNU 2.3% 192 St Richard's Hospital LNU 2.2% 642 Stoke Mandeville Hospital LNU 1.9% 453 Wexham Park Hospital LNU 1.0% 469 Oxford University Hospitals, John NICU 1,002 2.7% Radcliffe Hospital Princess Anne Hospital NICU 5.5% 872 Queen Alexandra Hospital NICU 2.9% 586 Trent Perinatal & Central Newborn Neonatal ODN George Eliot Hospital SCU 8.8% 255 Pilgrim Hospital SCU 11.5% 383 Queen's Hospital, Burton on Trent SCU 1.2% 387 Warwick Hospital SCU 7.1% 488 Kettering General Hospital LNU 4.9% 358 King's Mill Hospital LNU 5.1% 286 Lincoln County Hospital LNU 12.7% 677 Northampton General Hospital LNU 3.9% 589 Royal Derby Hospital LNU 5.6% 426 Leicester Neonatal Service2 NICU 8.9% 1,527 Nottingham City Hospital NICU 12.1% 931 Nottingham University Hospital (QMC) NICU 8.1% 810 University Hospital Coventry NICU 7.3% 1,867 Yorkshire & Humber Neonatal ODN Bassetlaw District General Hospital SCU 11.3% 173 Harrogate District Hospital SCU 11.7% 132 Scarborough General Hospital SCU 5.8% 226 Airedale General Hospital LNU 5.6% 222 Barnsley District General Hospital LNU 3.3% 348 Calderdale Royal Hospital LNU 0.8% 505 Chesterfield & North Derbyshire Royal 469 LNU 7.7% Hospital Diana Princess of Wales Hospital LNU 5.7% 1,173 Doncaster Royal Infirmary LNU 7.2% 511 Pinderfields General Hospital LNU 1.7% 925 Rotherham District General Hospital LNU 9.0% 394

2 Leicester Neonatal Service includes data from Leicester Royal Infirmary and Leicester General Hospital.

147

National Neonatal Audit Programme 2018 report on 2017 data

Scunthorpe General Hospital LNU 7.0% 668 York District Hospital LNU 3.5% 407 Bradford Royal Infirmary NICU 3.4% 842 Hull Royal Infirmary NICU 6.6% 1,155 Leeds Neonatal Service3 NICU 5.5% 1,652 The Jessop Wing, Sheffield NICU 1.8% 1,646 Isle of Man Nobles Hospital LNU 2.2% 118 SCOTLAND Borders General Hospital, Melrose SCU 23.3% 83 Dumfries & Galloway Royal Infirmary LNU 15.0% 189 Forth Valley Royal Hospital LNU 1.8% 375 Raigmore Hospital, Inverness LNU 8.4% 217 Royal Alexandra Hospital, Paisley LNU 9.8% 454 Aberdeen Maternity Hospital NICU 4.9% 843 Ayrshire Maternity Unit, Crosshouse NICU 5.4% 338 Ninewells Hospital, Dundee NICU 3.5% 529 Princess Royal Maternity, Glasgow NICU 2.0% 571 Royal Hospital for Children, Glasgow NICU 2.6% 999 Victoria Hospital, Fife NICU 8.5% 354 Wishaw General Hospital NICU 3.4% 749 WALES Ysbyty Gwynedd SCU 7.2% 226 Glan Clwyd Hospital LNU 6.3% 243 Glangwili General Hospital LNU4 3.3% 280 Nevill Hall Hospital LNU4 9.5% 213 Prince Charles Hospital LNU4 3% 269 Princess of Wales Hospital LNU4 5.9% 229 Royal Glamorgan Hospital LNU 2.5% 256 Wrexham Maelor Hospital LNU 15.2% 207 Royal Gwent Hospital NICU 3.9% 464 Singleton Hospital NICU 1.6% 464 University Hospital of Wales NICU 5.9% 501

3 Leeds Neonatal Service includes data from Leeds General Hospital and St James’ Hospital. 4 We are aware of designation changes in 2017 at these units, from LNUs to SCUs.

148

National Neonatal Audit Programme 2018 report on 2017 data

Appendix B: Recommendations by audience

The NNAP report makes a number of recommendations relating to the key findings from each audit measure. These recommendations are targeted wherever possible to a particular audience. Recommendations are listed by audience in the table below with recommendation number in brackets.

Recommendations for neonatal units and neonatal teams

The key findings and recommendations in this chapter are specific to each audit measure, however there are a number of recommendations for neonatal units relating to quality improvement activities across all NNAP measures:

2. Neonatal units should use NNAP Online to identify quality improvement opportunities relevant to them, and to identify partner units with results they wish to emulate.

3. Neonatal units should ensure they have adequate processes for the timely capture of information for quality improvement.

149

National Neonatal Audit Programme 2018 report on 2017 data

Antenatal (8) Neonatal units with below average rates of magnesium magnesium sulphate administration should identify comparable units to their sulphate own, that have higher rates of antenatal magnesium sulphate administration using NNAP Online. Working collaboratively with maternity staff, they should use quality improvement methodology and programmes to improve rates of administration in their hospitals.

(10) Neonatal units with significant levels of missing data should take steps to address this in collaboration with maternity care staff. Temperature on (15) Neonatal units should report all cases where the admission admission temperature of a very preterm baby is below 36.00C using local risk reporting mechanisms, and consider a policy of reporting all babies with admission temperature below 36.50C.

(16) Neonatal units should ensure that they have a care bundle in place, developed with multidisciplinary input, which mandates the use of evidence-based strategies to encourage admission normothermia of very preterm babies. Parental (17) Neonatal units should regularly review the reasons why consultation timely parental consultations did not occur. They should look for themes among the reasons, provide regular feedback to neonatal staff, and put processes in place to strengthen their support of parental partnership in care.

(18) Neonatal units should ensure that parents are aware of the standard, for example as part of a welcome pack or signage in the neonatal unit.

(19) Neonatal units with poorer data completeness should review and improve their documentation process. For example, by use of a dedicated notes sheet or a document in electronic records to record parental consultations. Parental presence (20) Neonatal units with poorer data completeness should review on consultant and improve their documentation process to ensure that all ward rounds instances of parental presence on the ward round are recorded.

(21) Neonatal units should work with local parent representatives to look at ways to improve the attendance of parents on the ward round and parental involvement in decision making. Neonatal units should refer to the BAPM Neonatal Service Quality Indicators34 and the Bliss Baby Charter8 for guidance.

150

National Neonatal Audit Programme 2018 report on 2017 data

On time ROP (22) Neonatal units and ophthalmologists should target quality screening improvement in their organisational, administrative and clinical processes at those babies whose birthweights and gestations are just inside the criteria for screening who constitute the majority of those not screened.

(23) Neonatal units with low outlier status, and especially those who have been recurrently identified as such, should urgently review their clinical, administrative and organisational arrangements, and keep them under detailed regular review to optimise retinopathy screening and treatment outcomes.

(24) Neonatal units should, as part of a formal local risk incident investigation, formally review their clinical, organisational and administrative pathways in discussion with their ophthalmology colleagues when cases are screened late, or not at all.

(25) Neonatal units should clearly describe to parents, prior to the opening of the screening window, but after the first week of life, the need for ROP screening using an individualised written resource which sets out for the parents the anticipated date of first screening for their baby. If their baby is due to be screened after being discharged from the unit, neonatal staff should ensure that parents are aware of the importance of attending the appointment. Encephalopathy (28) Neonatal units should ensure that all cases of encephalopathy identified by the criteria used by the NNAP have been reviewed by a suitable multidisciplinary group to look for modifiable factors, in accordance with the approach taken in the Royal College of Obstetricians and Gynaecologists’ “Each Baby Counts” programme35.

151

National Neonatal Audit Programme 2018 report on 2017 data

Measuring rates of (29) Neonatal units should enter every blood culture result that is infection positive for any bacterial or fungal growth (including potential contaminants) and use a regular communication channel with their laboratory services to assure themselves and other NNAP audit users that their data entry is complete.

(30) Neonatal units with complete entry of positive blood cultures and above average rates of bloodstream infection with a known pathogen in babies of less than 32 weeks gestational age should consider identifying suitable partner units from NNAP Online with lower rates of infection and comparing their infection reduction strategies to seek quality improvement opportunities.

(31) Neonatal units with complete entry of positive blood cultures and above average rates of bloodstream infection with central line use as measured by the CRG / NNAP quality improvement surveillance definition of central line associated bloodstream infection (QISD CLABSI) should consider identifying suitable partner units from NNAP Online, and comparing their infection reduction strategies to seek quality improvement opportunities. Bronchopulmonary (33) Neonatal units with a positive treatment effect should dysplasia consider examining the practice of neonatal units with a negative treatment effect to identify potential modifiable factors in their neonatal care which might influence rates of BPD.

(37) Neonatal units should ensure that they will be able to validate their NEC data entry for the 2018 data year.

152

National Neonatal Audit Programme 2018 report on 2017 data

Minimising (41) Neonatal units and trusts/health boards where transitional separation of care cannot be delivered should work with their commissioners mother and baby to develop the ability to deliver such care to minimise mother and baby separation, following the BAPM guidance A Framework for Neonatal Transitional Care.11

(42) Neonatal units with above average numbers of separation days for term, or late preterm babies should consider if revision of their admission or discharge criteria and processes could reduce the number of mother and baby separation days.

(43) Neonatal units should implement the BAPM guidance on the management of neonatal hypoglycaemia in term babies unless local circumstances make this inappropriate. Hypoglycaemia is a leading cause of term admission; some admissions for the management of hypoglycaemia could be avoided with the use of BAPM guidance.12

(44) Neonatal units should be aware of their rates of admission for term babies, and use the themes emerging from ATAIN project reviews in England of term admissions to inform possible targeted review of their admission and discharge processes. Breastmilk feeding (47) Neonatal units should use these data, alongside available at discharge home data concerning breastfeeding practices in non-admitted babies in their local area, to inform local quality improvement activity aiming to improve rates of breastmilk feeding. Neonatal units can use The Baby Friendly Initiative (UNICEF)14 and the Bliss Baby Charter8 to support this activity.

153

National Neonatal Audit Programme 2018 report on 2017 data

Follow up at two (49) Neonatal units and networks should adopt the NICE years of age guideline on Developmental follow-up of children and young people born preterm,33 and make progress with the implementation of pathways across organisational structures (e.g. Sustainability and transformation plan footprints in England). This requires a multidisciplinary, whole health economy approach.

(50) Neonatal units with incomplete data capture should ensure that they have the processes in place to document follow up at two years of age.

(51) Neonatal units should discuss arrangements for two-year follow up with families prior to discharge home of their baby, supported by written communication which includes the expected timeframe for the follow up consultation.

Recommendations for neonatal networks

Antenatal steroids (5) Neonatal networks should review administration rates of antenatal steroids in their units on a quarterly basis, identify any quality improvement opportunities and support units to achieve the best possible neonatal outcomes. Antenatal (11) Neonatal networks, maternity networks and local maternity magnesium systems with below average rates of administration, or low rates sulphate of improvement review administration rates of magnesium sulphate in their units on a quarterly basis, identify any quality improvement opportunities and support units to achieve the best possible neonatal outcomes. Birth in a centre (13) Neonatal networks should facilitate local and network review with a NICU of all cases where babies of less than 27 weeks gestational age deliver in a hospital without a NICU with the aim of identifying and sharing opportunities to increase the rate of delivery in hospitals with an onsite NICU.

(14) Neonatal networks, maternity networks and local maternity systems in England, and their equivalent bodies in Wales and Scotland, which do not achieve delivery of 85% of babies less than 27 weeks in a hospital with an onsite NICU should review whether they have realistic plans to achieve improvements in this area, and develop plans if required.

154

National Neonatal Audit Programme 2018 report on 2017 data

On time ROP (26) Neonatal networks with higher rates of failure to screen on screening time (for example over 2.5%) should seek to understand the reasons for this failure and address this with any units concerned.

Bronchopulmonary (34) Neonatal networks with a positive treatment effect should dysplasia consider examining the practice of networks with a negative treatment effect to identify potential modifiable factors in their neonatal care which might influence rates of BPD.

(35) When the NICE guidance on specialist neonatal respiratory care for babies born preterm is published, neonatal networks and neonatal units should review their policies to ensure that saturation targets are in line with best practice recommendations. Necrotising (38) Neonatal networks should support neonatal units providing enterocolitis all levels of care to undertake quality improvement activities relating to NEC. Minimising (45) Neonatal networks should work collaboratively with local separation of maternity system and maternity and neonatal safety mother and baby collaborative colleagues (or their equivalents in Scotland and Wales) to understand the themes emerging from the ATAIN project and to assist their units in reducing unnecessary separation of the mother and her term baby. Follow up at two (49) Neonatal units and networks should adopt the NICE years of age guideline on Developmental follow-up of children and young people born preterm,33 and make progress with the implementation of pathways across organisational structures (e.g. Sustainability Transformation Plan footprints in England). This requires a whole health economy and multidisciplinary team approach.

Recommendations for perinatal services

Antenatal steroids (4) Perinatal services (maternity and neonatal staff) should regard their rates of antenatal steroid administration as a key measure of the achievements of their clinical care. To identify quality improvement opportunities, neonatal and maternity care staff should formally review records of babies born at less than 35 weeks admitted for neonatal care where antenatal steroids were not given to the mother as part of their assurance with respect to NICE guidance.18

155

National Neonatal Audit Programme 2018 report on 2017 data

Antenatal (9) To seek missed opportunities, and themes as to why magnesium magnesium was not given in line with NICE guidance,18 neonatal sulphate and maternity care staff in units with below average rates of administration should formally review records of babies born at less than 30 weeks where magnesium sulphate was not given to the mother.

156

National Neonatal Audit Programme 2018 report on 2017 data

Recommendations for the NNAP and other national audits

Antenatal steroids (6) The NNAP and the National Maternity and Perinatal Audit should consider whether antenatal steroid administration could be more appropriately audited as part of the National Maternity and Perinatal Audit from 2019 onwards. Antenatal (12) The NNAP and NMPA should explore the feasibility of magnesium reporting antenatal magnesium administration in NMPA. sulphate

Measuring rates of (32) The NNAP should continue to seek to achieve linkage infection between other infection surveillance systems and the National Neonatal Research Database (NNRD) to report meaningful data about bloodstream infection. Necrotising (39) The NNAP should consider increasing the time period for enterocolitis reporting NEC, to a rolling period of three years to maximise the discriminatory power of this measure.

(40) The NNAP should consider reporting a combined outcome of NEC or death from the 2019 data year, and should consider applying a matching approach to facilitate comparisons of rates between different networks and units. Minimising (46) The NNAP should seek to present the number of admissions separation of and separation days alongside the number of births in each mother and baby gestational age category.

Breastmilk (48) The NNAP should develop a measure of early breastmilk feeding at feeding. discharge home

Recommendations for others

Antenatal steroids (7) Those responsible for defining national maternity datasets (NHS Digital in England) should ensure that antenatal steroid administration is captured as part of routine maternity data.

On time ROP (27) Guideline developers should take the successful deployment screening of on time post discharge screening into account when describing appropriate clinical practice for ROP screening.

157

National Neonatal Audit Programme 2018 report on 2017 data

Appendix C: Glossary and abbreviations

ATAIN Avoiding Term Admissions Into Neonatal units

BAPM British Association of Perinatal Medicine

BPD Bronchopulmonary dysplasia

COP Clinical Outcomes Publication

CQC Care Quality Commission

CSF Cerebrospinal fluid

EPR Electronic patient record

HQIP Healthcare Quality Improvement Partnership

HRG Healthcare resource group

Hyperthermia A body temperature more than 37.5°C

Hypothermia A body temperature less than 36.5°C

LNU Local neonatal unit

MBRRACE-UK Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK

MCN Managed clinical network

NCAB National Clinical Audit Benchmarking

NCAPOP National Clinical Audit and Patient Outcomes Programme

NDAU Neonatal Data Analysis Unit

NEC Necrotising enterocolitis

NHSE NHS England

NHSI NHS Improvement

NICE National Institute for Health and Care Excellence

NICU Neonatal intensive care unit

NMC Nursing and Midwifery Council

158

National Neonatal Audit Programme 2018 report on 2017 data

NMPA National Maternity and Perinatal Audit

NNAP National Neonatal Audit Programme

NNRD National Neonatal Research Database

NNU Neonatal unit

Normothermia A body temperature between 36.5°C and 37.5°C

ODN Operational delivery network

PICC Peripherally inserted central catheter

RCM Royal College of Midwives

RCOG Royal College of Obstetrics and Gynaecology

RCPCH Royal College of Paediatrics and Child Health

RCOphth Royal College of Ophthalmologists

ROP Retinopathy of prematurity

SCU Special care unit

UAC Umbilical arterial catheter

UVC Umbilical venous catheter

159

National Neonatal Audit Programme 2018 report on 2017 data

Appendix D: Useful resources The Royal College of Paediatrics and Child Health

The Royal College of Paediatrics and Child Health (RCPCH) was founded in 1996 and now has over 17,000 members across the world. We play a major role in postgraduate medical education, professional standards, research and policy.

The RCPCH has a number of useful resources, including:

• Courses and online learning www.rcpch.ac.uk/education/courses • Paediatric Care Online http://pcouk.org/ • Support for continuing professional development www.rcpch.ac.uk/education/continuing-professional-development • Medicines for Children https://www.rcpch.ac.uk/resources/medicines-children-information-parents- carers • MedsIQ: Sharing QI resources for paediatric medicine safety www.medsiq.org/

British Association for Perinatal Medicine

The British Association for Perinatal Medicine improves standards of perinatal care by supporting all those involved in perinatal care to optimise their skills and knowledge, promote high quality, safe and innovative practice, encourage research, and speak out for the needs of babies and their families.

Find out more about BAPM here: https://www.bapm.org/

Bliss

Bliss is a national charity for babies born premature or sick. It exists to give every baby born premature or sick in the UK the best chance of survival and quality of life. Bliss supports families, campaigns for change and supports professionals, and enables life- changing research.

Learn about Bliss here: https://www.bliss.org.uk

160

National Neonatal Audit Programme 2018 report on 2017 data

The Healthcare Quality Improvement Partnership

The Healthcare Quality Improvement Partnership (HQIP) is led by a consortium of the Academy of Medical Royal Colleges, the Royal College of Nursing and National Voices. Its aim is to promote quality improvement in patient outcomes, and in particular, to increase the impact that clinical audit, outcome review programmes and registries have on healthcare quality in England and Wales. HQIP holds the contract to commission, manage and develop the National Clinical Audit and Patient Outcomes Programme (NCAPOP), comprising around 40 projects covering care provided to people with a wide range of medical, surgical and mental health conditions. The programme is funded by NHS England, the Welsh Government and, with some individual projects, other devolved administrations and crown dependencies. www.hqip.org.uk/national-programmes.

Clinical Outcomes Publication

Clinical Outcomes Publication (COP) is an NHS England initiative, managed by HQIP, to publish quality measures at the level of individual consultant, team and unit level using national clinical audit and administrative data. The programme uses the platforms of MyNHS and NHS Choices and aims to maximise the availability and accessibility of outcomes and performance information to patients, the public and stakeholders.

National Clinical Audit Benchmarking

The National Clinical Audit Benchmarking (NCAB) initiative was originally created as a collaboration between HQIP and CQC, with a vision to enhance the way not just inspectors, but also medical directors, local clinical audit teams and others engage, interact with and share clinical audit data. The NCAB platform distils what can be necessarily complex reporting by national clinical audits into key metrics. Results are presented in an easy to understand visual form, specific for each Trust, hospital and in some cases ward, often against national benchmarks. These results are presented on an intuitive website platform (https://ncab.hqip.org.uk/), searchable by medical specialty or Trust/hospital/ward.

161

National Neonatal Audit Programme 2018 report on 2017 data

National Maternity and Perinatal Audit

The National Maternity and Perinatal Audit (NMPA) is a national clinical audit of NHS maternity services in England, Scotland and Wales. The audit, commissioned by HQIP, is led by the Royal College of Obstetricians and Gynaecologists in partnership with the Royal College of Midwives (RCM, the Royal College of Paediatrics and Child Health (RCPCH) and the London School of Hygiene and Tropical Medicine (LSHTM).

The audit evaluates a range of care processes and outcomes to identify good practice and areas for improvement in the care of women and babies.

You can find more information, access results and reports here: www.maternityaudit.org.uk.

Neonatal Data Analysis Unit

The Neonatal Data Analysis Unit (NDAU) at Imperial College London holds the National Neonatal Research Database (NNRD). The NNRD holds operational clinical information captured during care and supports health service evaluation and research. The NDAU analyse data for the NNAP.

Find out more about NDAU here: www.imperial.ac.uk/neonatal-data-analysis-unit.

162

National Neonatal Audit Programme 2018 report on 2017 data

Appendix E: Matching method of comparing outcomes for BPD

In a comparison of the outcomes of two groups that differ in their background profiles (the case-mix), two processes have to be considered:

A. Assignment of babies to units (how a mother/baby end up being treated/cared for in a particular unit)

and

B. How the unit treats babies

When we regard B. as important, and would like to discount A, a regression adjustment method applies a model for how the outcome depends on the treatment applied and on the background variables. When there are many background variables this method is inefficient - the standard errors are large. If the list of background variables is reduced, model selection introduces a bias in the estimates, especially when the selection involves many steps.

In a regression model, we are interested in the average effect associated with each network (or unit). This effect is defined for each individual baby, as well as for the network (or unit). A regression model assumes a particular pattern of the effects; in most cases, this assumption is that the network (or unit) has a ‘uniform’ (identical) effect on each baby. This assumption is not tenable - the effect is very likely to vary (substantially) across babies.

Fitting a regression model introduces several complications in addition to model selection. First, the analysis (and all the results) refer to the logit scale that does not have a straightforward translation to the probability (or percentage) scale.

The model becomes extremely complex as we introduce the networks as a factor, and the interactions of the networks with the most important background variables (e.g. gestational age). By its nature, model selection is obscure.

An impasse arises when a model that fits well cannot be found. A model cannot be declared in advance of the analysis. It is unlikely that the same model would be selected in the analyses for two consecutive years. By contrast, audit users might expect the same adjustment method to be applied in successive years when attempting to adjust an outcome for background variables. That the conclusions of an adjusted analysis (estimated treatment effects) would depend on the model that is selected, is a concern.

163

National Neonatal Audit Programme 2018 report on 2017 data

The principal arguments for analysis based on matching:

• The analysis adheres closely to the question that appeals to the clinical audience and is in accord with the purpose of the Audit: If our babies were treated elsewhere, would they have fared better? Or worse? • The matching approach separates the approach to the two key issues in comparing outcomes. It addresses assignment to networks/units and the quality of care provided by the network/unit in separate parts of the analysis: matching and analysis of the matched subgroups. • The outcomes are involved in the analysis only once, in the second part. • There are simple unambiguous criteria for successful matching - balance of the subgroups on all the background variables. • The small caseloads of numerous units raise no problems in the matching process, nor in the analysis that follows. • The analysis entails no bias that would arise from repeated use of the outcome variable. (In model selection, the outcomes are used in every model we test.) • The output from the matching exercise is a simple “treatment effect” percentage, (the difference between the matched rate and the estimated rate) and with an associated 95% confidence interval.

164

National Neonatal Audit Programme 2018 report on 2017 data

Appendix F: “Pathogens” in the NNAP

Bacterial, fungal and yeast positive blood cultures reported to the NNAP in 2017 for the Bloodstream infection and Central line associated bloodstream infection measures have been classified as shown below into organisms whose growth would be regarded as indicative of a bloodstream infection without further confirmatory evidence, and into a list of other organisms. This list of organisms included for NNAP reporting is available below. We are grateful to Professor Paul Heath* for reviewing and updating this list for the 2017 data year.

*Professor of Paediatric Infectious Diseases and Honorary Consultant, Paediatric Infectious Diseases Research Group; Director, St Georges Vaccine Institute.

Table B: Organisms included in NNAP reporting for the 2017 data year

Bacterial, fungal or yeast isolates indicating a clinically significant infection without additional data collection Acinetobacter sp. Enterobacter Pseudomonas aeruginosa agglomerans Acinetobacter Enterobacter cloacae Pseudomonas stutzeri baumannii Acinetobacter lwoffii Enterococcus sp. S. Aureus B haemolytic streptococci Enterococcus faecalis Salmonella sp. Bacillus cereus Enterococcus faecium Salmonella unnamed Beta-haemolytic strep. Escherichia coli Serratia sp. Group b Burkholderia capecia Flavobacterium sp. Serratia Candida sp. Gemella haemolysans Serratia liquefaciens Candida albicans Gemella morbilarum Serratia marcescens Candida dubliniensis Group b streptococcus Staphylococcus aureus Candida glabrata Haemophilus influenzae Staphylococcus capitis Candida guilliermondii Haemophilus Staphylococcus epidermidis parainfluenzae Candida kefyr Klebsiella sp. Staphylococcus haemolyticus Candida krusei Klebsiella aerogenes Staphylococcus hominis Candida parapsilosis Klebsiella oxytoca Staphylococcus warneri Candida tropicalis Klebsiella pneumoniae Stenotrophomonas maltophilia Citrobacter sp. Lactobacillus sp. Streptococcus agalactiae Citrobacter freundii Listeria sp. Streptococcus bovis Citrobacter koseri Listeria Streptococcus milleri monocytogenes

165

National Neonatal Audit Programme 2018 report on 2017 data

Clostridium perfringens Morganella morganii Streptococcus mitis Coliform Mrsa Streptococcus pneumoniae Corynebacterium Neisseria meningitidis Streptococcus viridans diphtheriae Eikenella corrodens Pasteurella multocida Treponema pallidum Enterobacter sp. Proteus mirabilis Yeasts Enterobacter aerogenes Pseudomonas sp. Organisms Acid fast bacilli Diphtheroids Rosemonas gilardii Actinomyces bovis Gram positive bacilli Scopulaiopsis brevicaulis Actinomyces sp. Haemophilus sp. Staph saprophyticus Aerococcus sp Kocuria species Staphylococcus sp. Staphylococcus - Aerococcus viridans Lactococcus sp. coagulase negative Staphylococcus - Alcaligenes faecalis Micrococcus sp. coagulase negative (mixed) Alpha haemolytic streptococci Mixed growth Staphylococcus simulans Stomatococcus Anaerobes Moraxella sp. mucilaginosus Bacillus sp. Moraxella catarrhalis Streptococcus sp. Bacteroides sp. Mycoplasma hominis Streptococcus - group g Chryseobacterium sp. Neisseria sp. Streptococcus anginosus Clostridium sp. Nocardia asteroides Streptococcus oralis Corynebacter Peptostreptococcus sp. Streptococcus salivarius Corynebacterium sp. Prevotella sp. Streptococcus sanguis Corynebacterium bacilli Propionibacterium sp Toxoplasma gondii Corynebacterium Proprionebacterium striatum acnes Ureaplasma Corynebacterium ulcerans Ralstonia sp. Yeasts (other) Gram positive cocci Shigella sonnei Group g streptococcus Sphingomonas

166

National Neonatal Audit Programme 2018 report on 2017 data

References

1 Wilson E., et al. Admission Hypothermia in Very Preterm Infants and Neonatal Mortality and Morbidity. The Journal of Pediatrics 2016; 175: 61-7.

2 lyu Y., et al. Association Between Admission Temperature and Mortality and Major Morbidity in Preterm Infants Born at Fewer Than 33 Weeks’ Gestation. JAMA Pediatrics 2015; 169e150277-8.

3 Laptook A.R., et al. Admission Temperature and Mortality and Morbidity among Moderately and Extremely Preterm Infants. The Journal of Pediatrics 2018; 192: 53-9.

4 Oddie S., Tuffnell D. J., McGuire W. Antenatal magnesium sulfate: Neuro-protection for preterm infants. Archives of Disease in Childhood - Fetal and Neonatal Edition 2015; 100: F553-F557.

5 NHS England. Implementing Better Births: Integrating Neonatal Care into Local Maternity System Transformation Plans. 2017.

6 Wilson E., et al. Admission Hypothermia in Very Preterm Infants and Neonatal Mortality and Morbidity. The Journal of Pediatrics 2016; 175: 61-7.

7 British Association of Perinatal Medicine. Neonatal Service Quality Indicators. 2017. Available from https://www.bapm.org/NSQI.

8 Bliss. Bliss Baby Charter. Available from https://www.bliss.org.uk/health- professionals/bliss-baby-charter.

9 Royal College of Obstetrics and Gynaecology. Each Baby Counts: 2015 Full Report. London: RCOG, 2017.

10 Stoll B.J., Hansen N.I., Adams-Chapman I., et al. Neurodevelopmental and Growth Impairment Among Extremely Low-Birth-Weight Infants With Neonatal Infection. JAMA 2004; 292(19): 2357–2365. doi:10.1001/jama.292.19.2357.

11 British Association of Perinatal Medicine. Neonatal Transitional Care – A Framework for Practice. 2017. Available from: https://www.bapm.org/resources/framework-neonatal- transitional-care

12 British Association of Perinatal Medicine. Identification and Management of Neonatal Hypoglycaemia in the Full Term Infant – A Framework for Practice. 2017. Available from: https://www.bapm.org/resources/identification-and-management-neonatal- hypoglycaemia-full-term-infant-%E2%80%93-framework-practice

13 Public Health England. Fingertips Tool. Available from https://fingertips.phe.org.uk/child-health-breastfeeding.

14 UNICEF. The Baby Friendly Initiative. Available from https://www.unicef.org.uk/babyfriendly.

15 Royal College of Paediatrics and Child Health. RCPCH Detection and Management of Outlier Status for Clinical Indicators for Managed Clinical Audits. 2017. Available from:

167

National Neonatal Audit Programme 2018 report on 2017 data

https://www.rcpch.ac.uk/sites/default/files/2018- 03/rcpch_detection_and_management_of_outlier_status_policy.pdf

16 Healthcare Quality Improvement Partnership. Detection and management of outliers for National Clinical Audits. 2017. Available from: https://www.hqip.org.uk/resource/detection-and-management-of-outliers-for- national-clinical-audits/

17 Scottish Government. The Best Start: A Five-Year Forward Plan for Maternity and Neonatal Care in Scotland. 2017. Available from http://www.gov.scot/Publications/2017/01/7728.

18 National Institute for Health and Care Excellence. Preterm labour and birth. NICE guideline (NG25) 2015. Available from https://www.nice.org.uk/guidance/ng25.

19 McDonald S. Antenatal corticosteroids for women at risk of preterm delivery. BMJ 2017; 356: j1467.

20 Smith G., Rowitch D., Mol B. The role of prenatal steroids at 34–36 weeks of gestation. Archives of Disease in Childhood - Fetal and Neonatal Edition 2017; 102: F284-F285.

21 Travers C., Clark R., Spitzer A., Das A., Garite T., Carlo W. Exposure to any antenatal corticosteroids and outcomes in preterm infants by gestational age: prospective cohort study. BMJ 2017; 356: j1039.

22 Oddie S., Tuffnell D. J., McGuire W. Antenatal magnesium sulfate: Neuro-protection for preterm infants. Archives of Disease in Childhood - Fetal and Neonatal Edition 2015; 100: F553-F557.

23 NHS England. Implementing Better Births: Integrating Neonatal Care into Local Maternity System Transformation Plans. 2017.

24 NHS England. Neonatal Critical Care Service Specification. 2016. Available from https://www.england.nhs.uk/commissioning/spec-services/npc-crg/group-e/e08/.

25 Scottish Government. Neonatal Care in Scotland: A Quality Framework. 2013. Available from http://www.gov.scot/Resource/0041/00415230.pdf.

26 Welsh Health Specialised Services Committee, NHS Wales. All Wales Neonatal Standards - 2nd Edition. 2013. Available from http://www.wales.nhs.uk/document/219405.

27 Department of Health. Toolkit for high quality neonatal services. 2009. Available from http://webarchive.nationalarchives.gov.uk/20130123200735/http://www.dh.gov.uk/en/ Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_107845.

28 Bliss. Bliss Family Friendly Accreditation Scheme. 2015. Available from https://www.bliss.org.uk/Shop/bliss-family-friendly-accreditation-scheme-booklet.

168

National Neonatal Audit Programme 2018 report on 2017 data

29 Royal College of Paediatrics and Child Health, Royal College of Opthalmologists, British Association of Perinatal Medicine, BLISS. Guideline for the Screening and Treatment of Retinopathy of Prematurity. 2008. Available from https://www.rcophth.ac.uk/wp- content/uploads/2014/12/2008-SCI-021-Guidelines-Retinopathy-of-Prematurity.pdf.

30 Stoll B.J., Hansen N.I., Adams-Chapman I., et al. Neurodevelopmental and Growth Impairment Among Extremely Low-Birth-Weight Infants With Neonatal Infection. JAMA 2004; 292(19): 2357–2365. doi:10.1001/jama.292.19.2357.

31 Isayama T., Lee S., Yang J., et al. Revisiting the Definition of Bronchopulmonary Dysplasia: Effect of Changing Panoply of Respiratory Support for Preterm Neonates. The Journal of Pediatrics 2017; 171(3): 271-279. doi:10.1001/jamapediatrics.2016.4141.

32 British Association of Perinatal Medicine. Neonatal Transitional Care – A Framework for Practice. 2017. Available from: https://www.bapm.org/resources/framework-neonatal- transitional-care

33 National Institute for Health and Care Excellence. Developmental follow-up of children and young people born preterm. NICE guideline (NG72) 2017. Available from https://www.nice.org.uk/guidance/ng72.

34 British Association of Perinatal Medicine. Neonatal Service Quality Indicators. 2017. Available from https://www.bapm.org/NSQI.

35 Royal College of Obstetrics and Gynaecology. Each Baby Counts: 2015 Full Report. London: RCOG, 2017.

169

National Neonatal Audit Programme (NNAP) 2018 annual report on 2017 data

Published by RCPCH September 2018

Healthcare Quality Improvement Partnership (HQIP) Dawson House, 5 Jewry Street, London EC3N 2EX

Royal College of Paediatrics RCPCH and Child Health Royal College of Paediatrics and Child Health 5-11 Theobalds Road, London, WC1X 8SH

The Royal College of Paediatrics and Child Health (RCPCH) is a registered charity in England and Wales (1057744) and in Scotland (SC038299).